DRAINAGE AND RECLAMATION OF SALT-AFFECTED SOILS
Transcript of DRAINAGE AND RECLAMATION OF SALT-AFFECTED SOILS
DRAINAGE AND RECLAMATION OF SALT-AFFECTED SOILS
CENTRALE LANDBOUWCATALOGUS
0000 0086 7438 B I R L I O T I I E E K
lANDEOUV...-^; . : .LU(&
Promotoren: dr. ir. W. H. VAN DER MOLEN, hoogleraar in de agrohydrologie dr. ir. P. BURINGH, hoogleraar in de tropische bodemkunde
hm c?zo 1 ^KP
J. MARTINEZ BELTRAN
DRAINAGE AND RECLAMATION OF SALT-AFFECTED SOILS IN THE
BARDENAS AREA, SPAIN
PROEFSCHRIFT
TER VERKRIJGING VAN DE GRAAD VAN DOCTOR IN DE LANDBOU WWETENSCHA PPEN,
OP GEZAG VAN DE RECTOR MAGNIFICUS DR. H. C. VAN DER PLAS,
HOOGLERAAR IN DE ORGANISCHE SCHEIKUNDE IN HET OPENBAAR TE VERDEDIGEN
OP VRIJDAG 27 OKTOBER 1978 DES NAMIDDAGS TE VIER UUR IN DE AULA
VAN DE LANDBOUWHOGESCHOOL TE WAGENINGEN
INTERNATIONAL INSTITUTE FOR LAND RECLAMATION AND IMPROVEMENT/ILR1 P.O. BOX 45, 6700 AA WAGENINGEN, THE NETHERLANDS, 1978
^
This thesis is also published as ILRI Publication no. 24
© International Institute for Land Reclamation and Improvement/ILRI, Wageningen, 1978
Printed in The Netherlands
/[//(/ 2 s7 o *) .„ i ./- -5<cy
PROPOSITIONS
The use of trickle irrigation can prevent the decrease in the acreage of
irrigated areas in the Canary Islands.
II
Since primary soil salinity is closely connected with geomorphology, the
physiographic approach is suitable for reconnaissance surveys -if one of
their objectives is to map saline soils.
Ill
The option of integration of Spain in the EEC appears to be the best
among the possible ones of joining other socio-economic areas, and the
option is favourable for Spanish agriculture in overall terms.
F. Botella. 1977. Situacion actual,
problematica y consecuencias de la
adhesion de Espana a la EEC. Revista
de estudios agrosociales no. 100.
IV
In the saline soils of the irrigated areas of the Ebro basin, sugar-beet
appears to be a more suitable reclamation crop than rice.
Crop rotations should be considered not only in terms of their economic
return, but also in terms of the suitability of crops in connection
with soil and water qualities and their effect on the salt balance.
Recommendation of an expert consult
ation on prognosis of salinity and
alkalinity. FAO. Soils Bulletin
no. 31. Rome. 1975.
VI
The successful long-term use of any irrigation water depends more on
rainfall, leaching, irrigation water management, salt tolerance of
crops, and soil management practices than upon water quality itself.
M. Fireman. 1960. Quality of water
for irrigation. University of Cal
ifornia Extension Service. Davis.
VII
The leaching efficiency coefficient depends not only on the irrigation
method and soil conditions but also on the subsurface drainage system.
VIII
It is more rewarding to invest in the preservation of the present land
resources under cultivation than to shift to new areas and draw on the
reserve of land resources.
Recommendation of an expert consult
ation on secondary salinization. FAO.
Rome. December 1977.
IX
The use of waste water for irrigation purposes appears to be advisable
in areas where water resources are scarce and where climate allows the
growth of early crops.
X
The increasing industrial expansion of a developing country has an
indirect effect on its agrarian reform.
Proefschrift J. MARTINEZ BELTRAN
Drainage and Reclamation of Salt Effected Soils in the Bardenas Area,
Spain.
Wageningen, 27 oktober 1978
A C K N O W L E D G E M E N T S
At the beginning of this book I would like to express my gratitude to everyone who made this publication possible.
I am most grateful to Prof. W. H. VAN DER MOLEN and Prof. P. BURINGH for their supervision, advice and discussions both in the field and in the University Departments.
I am greatly indebted to Dr. J. W. VAN HOORN for encouraging me to start with this thesis and for his contributions during the research work.
My sincere appreciation to Dr. N. A. DE RIDDER for his many valuable suggestions on the text and for his work as head of ILRI editorial committee.
I am most grateful to Ing. A. MARTINEZ BORQUE, former Director of the Instituto Nacional de Colonizacion, for his suggestion on the possibilities of reclaiming the saline soils of the Ebro irrigated areas, and for his constant support and encouragement.
I express my gratitude to Ing. T. VILLANUEVA ECHEVARRIA, Director of the Instituto de Reforma y Desarrollo Agrario, for his interest in the reclamation of the soils affected by salinity and for giving me the privilege of working in the experimental fields of the Bardenas area.
I am also indebted to Dr. J. BARDAJI CANDO, Head of the Soils Department of IRYDA, for his advice, especially on the geomorphology of the area, and for his valuable comments on the manuscript. Thanks are due to Mr G. GARCIA CRESPO and Mr. G. MORALES MURCIA for their collaboration in the field work of the reconnaissance survey, and to the staff of the soil laboratories of IRYDA in Madrid and Sevilla.
Special acknowledgement is due to Mr. A. P. WARDLE for his continuous advice and whose knowledge of the soil conditions of the Bardenas area was of great value. He also read the manuscript and corrected the English text.
The author is most grateful to Mr. U. CONDE BARREALES, whose accurate daily observations at the Alera experimental field made it possible to obtain the results presented in this book. The time spent by him in the field was a continuous lesson in practical agriculture. Thanks are also due to Mr. H. MARIN for his meteorological observations at the station of the Oliva Abbey.
The drawings were made by Mr. C. Ruiz whose work was most worthwhile. I am particularly grateful to Miss M. HERNANDO ANTON for her splendid work in typing the draft and to Dr. J. STRANSKY (ILRI) for typing the definitive text and for his work on the lay-out.
My sincere gratitude to Mrs M. F. L. WIERSMA-ROCHE for her comments on, and corrections in, the English text.
I wish to express my gratitude to Ir F. E. SCHULZE, Director of the International Institute for Land Reclamation and Improvement, and to the editorial committee for publishing the text in the ILRI series.
Finally, I wish to express my debt to my wife and children for the leisure time I spent, not with them, but in research work, and whose patience and generosity made this publication possible.
To my parents
Contents
INTRODUCTION
2. GENERAL CHARACTERISTICS OF THE AREA
2.1 Geology 3
2.2 Geomorphology 8
2.2.1 Drainage basins 8 2.2.2 Catchment area fringes 9 2.2.3 Main geomorphological processes 10 2.2.4 Major physiographic units 12
2.3 Climate
2.3.1 2.3.2 2.3.3 2.3.4 2.3.5
18
Temperature 19 Precipitation 20 Relative humidity Wind 25 Evapotranspiration
24
25
2.4 Natural vegetation 27
2.5 Present land use 30
2.5.1 Irrigated farming 30 2.5.2 Dry farming 31
3. HYDROLOGY 32
3.1 The irrigation and drainage systems 32
3.1.1 The irrigation scheme 32 3.1.2 The main drainage system 35
3.2 The groundwater table and drainage conditions 35
3.2.1 Rough mountainous land 38 3.2.2 Aragon alluvial valley 38 3.2.3 Rough eroded plain 39 3.2.4 Alluvial fans 40 3.2.5 Alluvial plain 40 3.2.6 Gypsum ridges and valleys 41 3.2.7 Mesas 41 3.2.8 Eroded mesas 42 3.2.9 Riguel alluvial plain 42 3.2.10 "Badlands", alluvial valleys and fans 43
SOILS AND SOIL CONDITIONS 45
4.1 General characteristics 45
4.1.1 Soil-forming processes 45 4.1.2 Parent materials 45 4.1.3 Topography 46 4.1.4 Land and water management 47
4.2 Mapping units 47
4.2.1 Mesas 52 4.2.2 Fluvio-colluvial valleys, slightly
saline-alkali phase 54 4.2.3 Piedmont slopes, slightly saline-alkali 56 4.2.4 Riguel river plain 57 4.2.5 High and middle Aragon terraces 59 4.2.6 Low Aragon terrace 62 4.2.7 Gypsum valleys 64 4.2.8 Stony ridges 64 4.2.9 Colluvial slopes 66 4.2.10 Fluvio-colluvial valleys, imperfectly drained
and saline-alkali phase 67 4.2.11 Alluvial plains, imperfectly drained
and saline-alkali phase 70 4.2.12 Alluvial valleys, saline-alkali phase 71 4.2.13 Alluvial fans, saline-alkali phase 74 4.2.14 Alluvial fans, slightly saline phase 75 4.2.15 Other minor mapping units 77
Soil profiles 1 to 17 78-111
5. SOIL SALINITY 112
5.1 Origin and localization of the salts 112
5.2 Types of salts and their distribution in the
soil profile 114
5.3 Salinity in the soil associations 115
5.4 Local crop tolerance to salinity and alkalinity 125
5.5 Soil salinity and land reclamation 129
DRAINAGE AND RECLAMATION EXPERIMENTAL FIELDS 132
6.1 Objectives of the field experiments 132
6.2 Selection criteria 133
6.3 Detailed soil survey 134
6.3.1 Outline of the survey 134 6.3.2 Hydrological soil properties 135 6.3.3 Initial salinity 142
7. EXPERIMENTAL FIELD DESIGN 148
7.1 Design of the drainage system 148
7.1.1 Alera experimental field 148 7.1.2 Valarena experimental field 162
7.2 Design of the irrigation system 168
7.2.1 Alera experimental field '68
7.2.2 Valarena experimental field 170
7.3 Soil improvement 170
7.4 Field observation network 170
7.5 Field observation programme 176
DERIVATION OF AGRO-HYDROLOGICAL FACTORS 177
1 Groundwater flow conditions 177
8.1.1 Alera experimental field 177 8.1.2 Valarena experimental field 18!
8.2 Drainage equations 183
8.2.1 Steady flow 183 8.2.2 Unsteady flow 184
,3 Determination of the drainable pore space 190
8.4 Determination of the hydraulic conductivity 193
8.4.1 Unsteady flow: Boussinesq equation 193 8.4.2 Steady flow: Hooghoudt equation 202 8.4.3 Hydraulic conductivity of the less
pervious layer 203 8.4.4 Comparison of the results obtained 205
8.5 Determination of the entrance resistance 207
8.5.1 Direct method of determining the W -value 208
e 8.5.2 Indirect method of determining the
W -value 214
9. DESALINIZATION PROCESS 220
9.1 Stages of the process 220
9.2 Water balance of the initial leaching phase 221
9.3 Theoretical leaching approach 229
9.3.1 Series of reservoirs 229 9.3.2 Numerical method 234
9.4 Actual desalinization process 235
9.5 Determination of the leaching efficiency
coefficient 244
9.6 Prediction of the initial leaching requirements 252
9.7 Leaching requirement during the cropping stage 254
10. SUBSURFACE DRAINAGE SYSTEM 258
10.1 Drainage criteria for unsteady-state conditions 258
10.1.1 Relation between crop yields and water table depth 259
10.1.2 Relation between water table depth and mobility of agricultural machinery 267
10.1.3 Rainfall distribution and water table rise 269
10.1.4 Relation between percolation losses from irrigation and water table rise 270
10.1.5 Assessment of drainage criteria for unsteady-state conditions 271
10.2. Drain spacing 273
10.3 Drainage criteria for steady-state conditions 274
10.4 Drain spacings required for different combinations of drainage and filter materials 276
10.4.1 Steady flow 276 10.4.2 Unsteady flow 278
LIST OF SYMBOLS 282
LITERATURE 285
SUMMARY 289
RESUMEN 299
SAMENVATTING 311
CURRICULUM VITAE 322
xii
1. Introduction
The Ebro basin is situated in north-eastern Spain and forms a geo
graphic unit that is bounded to the north by the Pyrenees, to the east
by the Catalan Coastal Mountains, and to the south-west by the Spanish 2
Plateau. It covers about 85 500 km .
The basin is drained by the Ebro river and its tributaries, which
rise in the Pyrenees and the Iberian Mountains.
The region is the driest part of northern Spain because it is
surrounded by high mountain ridges which isolate it from any Atlantic and
Mediterranean influences. If water is available, however, the climate
permits irrigated agriculture, the main crops being maize, lucerne, and
fruit trees.
The oldest irrigation schemes consist of canals diverted from the
Ebro river. Construction of the Canal de Tauste on the left bank com
menced in 1252 and the system of the Canal Imperial de Aragon,
which irrigates the right bank downstream of Tudela, was finished in
1786.
Early in the 20th century new irrigation plans were made to use the
surface water resources from the Pyrenees for agriculture. The Canal de
Aragon y Cataluiia and the Canal de Urgel were first constructed and some
years later the Bardenas-Alto Aragon irrigation scheme was designed.
This irrigation scheme consists of a series of reservoirs which will
regulate the flow of the three main tributaries of the Ebro river along
its middle reach, i.e. Aragon, Gallego and Cinca, and three main irrig
ation canals. Although parts of these canals had previously been constructed,
1
it was only after the Civil War that the Instituto Nacional de Colonizacion
began the full transformation of the areas under irrigation command. When
the scheme is finally completed, about 276 000 hectares will be irrigated.
The first part of the Bardenas area forms part of the Bardenas-Alto
Aragon irrigation scheme. This area covers approximately 57 000 hectares.
It is irrigated with water from Bardenas canal, which receives its supplies
from the Yssa reservoir in the Aragon river.
Within the area under command of the Bardenas canal, soil units
affected by salinity occur. To study the possibilities of reclaiming the
land for irrigated agriculture, the Soils Department of the Instituto
Nacional de Reforma y Desarrollo Agrario conducted a reconnaissance
survey in the area between 1972 and 1974. During the survey the saline
soils and those wi^h ii risk of secondary salinization were mapped and a
study was ma.it; of the causes of soil salinization.
One or the conclusions of the soil survey was that a field experiment
should be conducted LO study the technical and economic aspects of drain
ing and leaching these saline soils. In 1974, therefore, following a
detailed hydropedoi.ogical survey, two drainage and desalinization experimen
tal fields were, selected, in these pilot fields the drainage system, the
leaching process, an-I the land and water management required to prevent
resalinization were studied from 1974 to 1978.
2. General characteristics of the area
2.1 Geology
The Bardenas area forms part of the Ebro basin, which is bounded to
the north by the Pyrenees, to the east by the Catalan Coastal Mountains,
and to the south-west by the Spanish Plateau (Fig.2.1).
fc*;."j Quaternary -Recent alluvium /
Post-Alpine grounds \
Pre -Aipine grounds
Boundary of the basin
2 0 0 20 40 6 0 80 lOOKm
Fig. 2.1: Simplified geology of the Ebro basin (from the Geological Map of Spain 1 : 1 000 000).
The origin of the salinity that affects some of the soils in the area
is related to the geological processes involved in the formation of the
Ebro basin.
During the Secondary Era the present Ebro basin was a sea arm between
the Mediterranean and the Atlantic. In that period transgressions with
marine deposition and regressions with continental deposition alternated.
When the Cenozoic period started, Alpine folding affected the basin
structure. During the first folding phase both the Pyrenees and the Cat
alan Coastal Mountains emerged, the Iberian Plateau arose, and the former
sea arm was closed and became a great Oligocene lake of brackish water.
During the Tertiary Era dry and wet periods alternated. Fine sediments
were deposited in a brackish environment during periods of high evaporation
and evaporites were locally formed. Meanwhile during wet periods coarser
sediments were deposited near the borders of the basin.
During the late-Pliocene/early Pleistocene a system of ancient water
courses carried coarse sediments from the basin border into the area. This
coarse alluvium was deposited on the finer lacustrine sediments. Mesas of
coarse mixed alluvium were formed (Fig.2.2).
Finally the present Ebro River was formed after a breakthrough of
the Catalan Mountains. As a result the former Oligocene lake was drained.
During the Quaternary the base level was considerably lowered. The
Ebro river and its main tributaries thus incised and formed alluvial
valleys in the Tertiary sediments.
In the Bardenas area the lacustrine sediments comprise mudstones with
interbedded layers of siltstones (Fig.2.3). In the south fine limestone
layers overlie the mudstones and in the west gypsum strata are interbedded
in the mudstone layers (Fig.2.4).
ijjjf>i& '"•*"
. - '-0 x
Fig.2.2: Coarse mixed alluvium (mesa).
Fig.2.3: Mudstones with interbedded layers of siltstones (Bardenas canal).
Copdrroso'
•V-. W *~~ v^) f Ejeo de los
Coballeros
LEGEND
ALLUVIAL.-Clay and gravel
r-V"-°l PLIOCENE -PLEISTOCENE.- Conglomerate
UPPER MIOCENE - Limestone and mudstone
LOWER MIOCENE- Limestone and mudstone
LOWER MIOCENE.- Mudstone and siltstone
LOWER MIOCENE.- Conglomerate
OLIGOCENE.- Mudstone,gypsum and siltstone
Boundary of catchment areas
5Km 0 L. ^ u
lOKm
Fig. 2.4: Geology of the Bardenas area (simplified from the Geological Map of Spain 1 : 200 000).
Where the sediments contain thick layers of gypsum, folding has
occurred due to diapiric processes by which the more plastic evaporites were
deformed under the weight of the overlying strata (Fig.2.5).
Fig. 2.5: Folded gypsum strata interbedded in saline mudstones (road Zaragoza - Pamplona).
Two points are important in regard to soil formation. The first is
that the siltstones consist of silt and clay-sized particles cemented by
calcium carbonate. The second is that both mudstones and siltstones con
tain soluble salts because they were formed in a brackish environment. This
means that the soil parent materials derived from these rocks are fine-
textured with a high content of clay and silt, and a low content of fine
sand. They are also saline. Other soil parent materials derived from coarser
7
sediments, e.g. the mesas and the river terraces, have a coarser texture and
are relatively free of salts.
2.2 Geomorphology
During the Quaternary period the main landscape forming process was
erosion, followed by transport and local deposition of material. The pre
dominance of erosion, together with a gradual lowering of the base level,
has resulted in the oldest Miocene-Oligocene formations occupying the
highest points in the landscape, thus forming the uplands of a dissected
plain. At a lower level are the mesas, which consist of partially eroded
lacustrine sediments covered by younger deposits of coarse gravel. Sub
sequent erosion has removed large parts of the mesas and has cut deeply
into the underlying lacustrine formations.
Most of the eroded sediments were removed from the area, but local
sedimentation also occurred. In this semi-arid area both sediments and
salts moved together and were even deposited together. The highest salt
concentrations are therefore found in the lowest parts of the alluvial
formations where the alluvium was derived from the eroded mudstones and
siltstones.
To understand the origin of the salinity and its distribution and
intensity, one must carefully analyse the main physiographic units and
the processes involved in the landscape formation. Once a relationship is
established between soil salinity and the physiographic units, one has
a sound basis on which to map the saline soils.
2.2.1 Drainage basins
Within the project area two catchment areas can be distinguished
(Fig.2.4). The northern part drains towards the Aragon river, which is
deeply incised into the Miocene-Oligocene sediments in the north. The
river changes its course westwards at Carcastillo and below this point has
formed a broad alluvial valley.
8
The southern part of the area is drained by the Riguel river, which comes
from the north-east and flows through the area from north to south. The
Riguel debouches into the Arba de Luesia river at the southern end of the
area.
Both the Aragon and the Arba are tributaries of the Ebro. The first
flows into the Ebro near Milagro 70 km upstream from Gallur, where the
Arba joins the Ebro.
This means that the base levels of the two valleys differ by an
altitude of some 70 m. This has important consequences for the geomorphology
of the area, the Riguel basin being far more deeply eroded and more dissected
than the Aragon valley.
2.2.2 Catchment area fringes
a) The Arag6n basin. Though the whole northern part of
the project area drains towards the Aragon river, only a section of the
north-western part does so directly. The rest of the basin is virtually
isolated because it is almost surrounded by Miocene-Oligocene uplands.
The eastern border of the basin is the mesa de Larrate which, apart
from the uplands, is the highest feature of the northern area. To the south
some residual relief forms the interfluve between the Aragon and the Riguel
basins. The Castiliscar stream has cut the eastern mesa into two parts and
through this gap the basin drains towards the Aragon.
The north-western end of the area is bounded to the south by a high
mesa, while to the west the watershed runs along an Oligocene mudstone and
gypsum formation.
b) The Riguel and Arba basin. Uplands of Miocene sediments
form the western fringe of the Riguel catchment area. These sediments
consist of mudstones and siltstones, while in the south higher Miocene
limestone occurs. The river has cut up an ancient Quaternary mesa and
has formed an alluvial valley in between. The interfluve between the
Riguel and the Arba catchment areas runs along the eastern edge of the
mesa. Only a very small part of the studied area drains directly into
the Arba.
2.2.3 Main geomorphological processes
The evolution of the present landscape is the result of erosion
working upon the Miocene-Oligocene sediments and the ancient Quaternary
mesas, plus deposition of sediments transported by streams. Thus erosion
followed by transport and deposition under semi-arid climatic conditions
is the main process involved.
In the northern basin a former rough Miocene plain of mudstones and
siltstones was eroded during the Quaternary. As a result some residual
relief remains wherever a hard siltstone layer has protected softer mud-
stone underneath. Denudation and transport of coarse materials from eastern
uplands during the early Quaternary formed local colluvium that was after
wards eroded again. Of this coarse alluvium, stony ridges remain as resi
dual relief.
Further erosion-deposition processes have formed narrow fluvio-colluvial
valleys, which run westwards between the residual relief. Major streams
coming from the surrounding uplands have formed alluvial fans. These fans
have not developed their characteristic triangular shape because their
deposition has been restricted by the presence of many residual hills.
Near Alera several fluvio-colluvial valleys merge into a flat alluvial
plain which is surrounded by higher residual hills. As the plain outlet is
rather narrow, both the lowest part of the fluvio-colluvial valleys and the
alluvial plain itself are poorly drained.
Between the residual hills and the alluvial formations, eroded slopes
mixed with local colluvium prevail.
10
In the north-western part of the project area the main geomorphological
process has been the deposition of mixed alluvium by the Aragon river during
the Quaternary inter-glacial periods and erosion during the glacial epochs.
Sedimentation still continues and so mixed alluvial land is being formed
in the present flood plain. The result of these processes is the formation
of a broad alluvial valley in which three levels of fluvial terraces can
be distinguished.
Between the fluvial terraces and the higher conglomerate mesa, a
piedmont slope occurs which is now being eroded.
In the north-western end of the area, erosion working upon the tilted
Oligocene gypsum sediments has resulted in parallel gypsum-ridges and
associated valleys.
In the Riguel basin erosion was deeper and has worked upon the Miocene
sediments and upon the remaining mesas as well. Between the mesas and the
Riguel alluvial valley a piedmont slope occurs. Erosion is now working
upon the soft mudstone of the slope and upon the harder conglomerate of
the mesa escarpment. The result is the formation of eroded slopes and narrow
erosion valleys which are cutting the mesa edges. Due to this process some
isolated mesas remain as residual hills.
In the western part of the area soft mudstones and some parts of the
remaining old colluvial ridges have been eroded. Denudation of nearby
Miocene uplands and further transport and deposition of fine saline allu
vium by streams have formed narrow alluvial valleys and fans.
The weathered mudstones and siltstones still contain salts. The
alluvium derived from these Miocene-Oligocene sediments also contains so
luble salts as in this semi-arid climate no desalinization occurred during
its transport. This alluvium is even more salty than the weathered mudstones
themselves, and the further it has been transported the higher is its salt
content.
On the other hand, the ancient alluvium of coarse fragments covering
the Pliocene-Pleistocene mesas and the mixed alluvium deposited by the
11
Aragon and Riguel rivers are non-saline because their denudation areas
are outside the Miocene-Oligocene saline sediments. Moreover these forma
tions have a higher permeability and during rainy periods have lost any
salts initially present.
2.2.4 Major physiographic units
Within the whole project area the major physiographic units were
defined (Fig.2.6). Each of these is subdivided into minor components indi
cated on the physiographic soil map (Chap.4). The relative position and
the relationship between the different physiographic units are illustrated
in schematic cross sections (Figs. 2.7 through 2.12).
Of the units described 1-6 are confined to the Aragon drainage basin
and therefore only occur in the northern part of the area. Unit 7 is
common to both, whereas units 8-10 are only found in the Riguel-Arba basin.
1. Rough mountainous land. This is formed by the dissected
Miocene-Oligocene uplands surrounding the area. It appears inside the area
only near the northern border.
2. Arag6n alluvial valley. This comprises three fluvial terra
ces and a recent flood plain of mixed alluvial land (Fig.2.7). The diffe
rent terraces are separated by terrace escarpments. Medium and high terra
ces consist of coarse alluvium and the lower terrace of alluvial sands. All
terraces are covered by medium textured sediments no thicker than 1 m.
The thickness of the alluvium above the mudstone is more than 3 m in all
terraces. The lower terrace is about 325 m above mean sea level, the medium
terrace about 360 m, and the higher terrace about 370 m.
3. Rough eroded plain. This unit has a highly irregular surface
consisting of the residual relief of a Miocene plain which was eroded
during the Quaternary (Fig.2.8). It comprises minor physiographic units
as residual stony ridges and siltstone outcrops, eroded slopes with local
colluvium, and intervening fluvio-colluvial valleys. In these valleys
12
Coporroso
Fig.2. 6: Map of major physiographic units.
deposits are found which are derived from the surrounding slopes plus local
alluvium laid down by small streams.
4. Alluvial fans. In the northern basin some ephemeral streams
which flow from the surrounding uplands have formed alluvial fans where
they enter the rough eroded plain.
13
Piedmont Aragtin fluvial valley | slope) Mesa
2.1 ,13.11 3.2 12.1 |3.3| | 3.4 |22|2I| 1.1
_£.
-450
400 ~
350
-300
Fig.2.7: Schematic cross-section of the Aragon alluvial valley.
Aragon I vodey | M e s a Piedmont slope Rough eroded I Aragon Riguel
interfluve
3.1 | | 1.1 | | 2.2 M 3 - 8 ! 2 6 II 2 6 I 3 > 2 I 2 4 | | 2.4 | 3.8 | | 24 14
Pardina -450 irrigation conal
Fig.2.8: Schematic cross-section of the rough eroded plain.
5. Alluvial plain. Several fluvio-colluvial valleys coalesce
to form a poorly drained alluvial plain. This was formerly a seasonally
swampy area composed of relatively recent alluvium which has now been
drained by improving and deepening the natural outlets.
6. Gypsum ridges and valleys. In the north-western end of the
project area a distinct unit is formed by long parallel gypsum ridges and
narrow valleys (Fig.2.9). They are the consequences of the erosion of the
tilted gypsiferous sediments of the area. The ridges form steep escarpments
facing south.
14
Aragon fluvial valley| Gypsum ridges and valleys |Uplan
21 2.1 3.1 |3 .2 | i 3.3 I I I
Aragon river
|17| 3.7 I 1.7 |3.7| 1.7 I 2.71 2.4 I 1.3 -400 »
350
300
Fig.2.9: Schematic cross-section of the gypsum ridges and valleys.
7. Mesas. With the exception of the rough mountainous land the
mesas, locally named "sasos", form the highest physiographic unit of the
project area. This unit is very extensive within the area and occurs in both
drainage basins.
These mesas are high old alluvial plains formed in late-Pliocene/
early-Pleistocene by deposition of coarse alluvium transported by water
courses flowing down from the Pyrenees in times of heavy rainfall (Figs.
2.10 and 2.11).
Rough eroded plain
l2.2'|2.l| 1.1 Sardenas -rigqtion anal
400 S
L 3 0 0
2 Km.
Fig. 2.10: Schematic cross-section of the studied area showing the Riguel river alluvial plain and two levels of mesas.
15
Bodlond and alluvial valley t I Eroded I Riguel river l E r o d e d mesn and Ions I mesa I olluviol plain I t r o d e a meso
Z* 25 2.5 |2.4|3J3| |l.6|2.4| 1.2 J24| 3.11 |2.4]1.1| | 1.1 | | 3.6 |2<|l.l| 3.9 |l2|za| 1.1 |25| 1.1 | " | 2 . 4 |
Cinco Villas
Fig.2.11: Schematic cross-section of the studied area at a lower level than that of Fig. 2.10.
The mean altitude of the mesas is 450 m above sea level, though eroded
mesas occur in lower positions. They are characterized by moderately shallow
red soils overlying calcified well-rounded coarse alluvium which in turn
overlies the Miocene mudstone.
The mesas are flat and their slope is generally less than 1 per cent
but always in north-south direction.
8. Eroded mesas. Together with the gradual lowering of the base
level of the Riguel river the mesa edges have been deeply incised by
temporary streams (Figs.2.10 and 2.11). Once the harder conglomerate has
been removed erosion working upon the softer mudstone becomes more inten
sive. As a result of this process the mesa edges are severely eroded and
only small isolated remnants of the major unit remain.
A transition between the mesas and the lower alluvial land is formed
by piedmont slopes, which consist of a mixture of local colluvium from
the higher ground and Miocene mudstone.
9. Riguel alluvial plain. Between the two eroded parts of the
ancient uniform mesa the Riguel river has laid down its own deposits
forming a broad alluvial plain (Figs.2.10 and 2.11). This alluvium is
more than 3 m thick and consists of fine sediments which overlie the
Miocene mudstone. These sediments are derived from outside the mudstone
area and are consequently free of salts.
At the moment the Riguel is entrenched in its own sediments and a
narrow levee of sandy loam soils can be distinguished.
16
10. Badlands, alluvial valleys, and fans. This unit is
found between the western mudstone and siltstone uplands and the western
edge of the eroded mesa (Fig.2.11). All temporary streams that transport
the eroded material from the uplands drain towards a permanent stream
which joins the Riguel in the south of the area. This Valarena stream
runs close to the mesa edge.
This major unit comprises the residual relief which remains as evi
dence of the erosion of the mudstones and old colluvial ridges composed
of angular cobbles. The unit also comprises narrow alluvial valleys and
in the south of the area alluvial fans whose lower parts reach the Riguel
and Arba valleys (Fig.2.12).
Miocene upland j Badland and alluvial valleys and fans
I 3.13 I 1.6 I 3.13 Cinco Villas irrigation canal
Al luvial plain
I 3.6 | Arba river
r-400
350 £
300
2 Km.
Fig.2.12: Schematic cross-seation of the southern piedmont unit.
The result of erosion and alluvial deposition is a "Badland" of re
sidual stony ridges and eroded slopes with intervening valleys and fans.
The alluvium of valleys and fans is quite similar and consists of
very fine saline material, very finely stratified, which locally overlies
a coarse alluvium of rounded gravel which in turn overlies the Miocene
impervious mudstone.
17
As the main object of the reconnaissance survey was to map and investi
gate the salt-affected soils and as there is a close relationship between
soil salinity and physiographic units, these units form the main soil
landscapes and broad soil associations. Every major unit comprises minor
physiographic units whose soils have a well-defined level of salinity.
The main physiographic soil units may therefore be regarded as the
basis for the highest level of soil classification for salinity mapping.
The physiographic soil map is in fact a soil association map which shows
the distribution of the units over the project area.
2.3 Climate
The Ebro basin is the driest part of northern Spain because it is
surrounded by high mountain ridges which isolate the region from any Atlantic
and Mediterranean influences. The centre of aridity is sited around Zarago-
za (Fig.2.13).
150 200 Km.
Fig. 2.13: Mean annual precipitation deficit in the Ebro basin (from Thornth-waite water balance). From F.Elias and R.Gimenez Ortiz, Madrid 1965).
18
By the Koppen classification the climate of the Bardenas area is tem
perate with hot summers and evenly distributed rainfall (Cfa). By the
Thornthwaite classification the climate is semi-arid with a slight excess
of water in winter only in the north of the area (DB' ).
From north to south the climate becomes drier. There is a marked
seasonal temperature variation with hot summers and cold winters. Evapo-
transpiration exceeds total precipitation; the latter is extremely variable
and is not concentrated in distinct rainy seasons, although summer and
winter are slightly drier than autumn and spring. Wind velocity is high
mainly in winter and spring, and both cold and warm dry winds are common.
Climatological data have been collected from two observation stations,
one in the north (La Oliva) and one in the south (Gallur). The data are
compiled in Table 2.1.
2.3.1 Temperature
Figure 2.14 shows a great variation of mean temperature over the year.
The average values range between 5 C in winter and 23 C in summer and
there is 1 C mean temperature difference from north to south. In summer
day temperatures will sometimes reach 40 C and in winter minimum tempera
tures as low as - 7 C have been recorded.
The frost-free period is about 165 days and the growing season for
irrigated summer crops starts in early April and finishes in late October.
It is therefore long enough for maize but too short for rice. Night frosts
in early spring and frequent strong winds make fruit-tree cultivation
difficult in most parts of the area.
19
TABLE 2.1 Climatological data from Bardenas area observation stations (kindly provided by the National Meteorological Service)
Sta. N ° - ° f J F years
Mean temperature in C
1 36 4.3 5.8 2 18 5.5 6.9
M
9.4 10.6
A
12.2 13.0
H
15.3 17.4
J
19.6 21.1
J
22.5 22.7
A
22.1 23.2
S
19.1 20.3
0
14.3 14.8
N
8.5 9.4
D
5.3 6.6
Year
13 14
Extreme maximum temperature in C
36 18
15.7 15.5
17.5 17.2
22.6 22.8
25.6 25.2
31.0 30.4
35.8 34.7
38.5 36.7
37.6 35.2
33.6 31.7
26.9 25.3
20.0 19.3
15.2 15.3
38.5 36.7
Mean maximum temperature in C
8.9 9.0
11.0 11.3
15.5 16.0
18.5 18.5
22.2 23.7
27.0 27.6
30.7 28.7
30.0 29.4
26.3 25.8
20.4 19.6
13.6 13.4
9.3 9.8
19.5 19.4
Mean minimum temperature in C
36 18
-0.4 1.9
0.5 2.5
3.2 5.2
5.9 7.4
12.1 14.5
14.2 14.1 16.6 16.9
11.9 14.8
3.3 5.4
1.2 3.3
Extreme minimum temperature in C
Mean relative humidity in percentage
3 56 75 68 62
Mean precipitation in i
39 18
35 23
28 20
33 25
52 33
45 47
Mean number of days with precipitation of 0.I mm or more
39 18
22 30 17 15
4 4 4 4
44 41
51 30
45 31
Mean number of days with snow on the ground
1 39 1.1 0.9 0.2 2 18 0.9 0.7 0.3
Wind speed in m. sec
3 56 4.2 4.5 4.5 4.8
Mean sunshine duration in hours per day
3 56 4.4 5.9 6.4 7.7
0.1 0.1
0.5 1.0
4.3 4 . 0
11.7 10.5
6.9 9.1
1 2
36 18
-7.0 -3.3
-5.8 -3.5
-2.8 0.6
-0.5 2.1
2.3 5.5
6.7 9.7
9.4 12.0
8.9 12.1
5.8 9.7
1.3 3.9
-3.2 1.4
-5.2 2.9
-7.0 -3.5
475 350
85 76
2.8 3.0
NOTE: Station 1: La Oliva 42°22' - 1°27' « 342 m Station 2: Gallia' 41°52' - 1°19' W 257 m Station 3: Zavagoza 41°39' - 53' W 237 m
2.3.2 Precipitation
Annual t o t a l p r ec ip i t a t ion decreases southwards towards the centre
of the Ebro basin. Mean values vary between 475 mm in the northern area and
350 mm a t the southern border.
20
__P_„H—I-
i i 1 ! ! I
; ; ! 1 ; ;
i 1
; i
•••-; | t -
....L-..-J
-+i /.,'
/ / i J
. zzrr-. ?%/£•••••• -..^.:-
_ : . ^ T T i l i i
—r"—•—! '! *—• • .— T ~ r~^-"~T— i ! : • . !
Z t i H - ^ U ^ ^ t - . - •^ - • :Zt Z\ M l fe=£ :; -Z
~x$£±:is±x~ i Z j T ^ - T ^ C ^ ^ j I ^ - i - -+l'fi:^,\r^^ ^_ T^G^^Xlpa: ^ •//Xi-i- ! r U-U- j
/ / 1 1 \ J •
2^= T Sfc^ rz^ .r. : .IT :-T r TI r , r ^ b - • &_
7t^~^ _/ l v~ • • j : i !;:- V
TT i4 - i - t - : -3Si_ '•II—L- -t : ^ 1 :-: 1 \ •:••"•>• •':;: ~nnnLi±i~t:- •= I u I 3 •••r:-:^ • •'. : . - i
"•: - 1 " ! • :'.'
—i—*—i—'—i—'—i—'—i—"—i—'—i—1 i ' 1 '
_. . . . Lo Oliva _x x x Gallur
Fig.2.14: Mean temperatures in C.
Total rainfall and its distribution differ from year to year, but
continuous periods of heavy rainfall are uncommon. The total amount
is well distributed over the year and there is no seasonal concentration of
precipitation. Both factors have their influence on the salt regime of
the soils.
For drainage design purposes a maximum rainfall analysis was made.
Depth duration frequency diagrams were drawn, using the Gumbel probability
distribution (Figs.2.15 and 2.16). From the data of mean precipitation
(Table 2.1) and Figs. 2.15 and 2.16, it could be concluded that in areas
of lower total yearly precipitation heavy rainfall intensity is higher. The
graphs show that periods of continued heavy rainfall are uncommon and there
is hardly any difference between the amount of precipitation expected in
3 and 6 consecutive days.
21
P Jf. -~-j mm.
"" •:-:::: - H : : . . .
. . L::: ! 1 :• • M
••.. : ; i :::: ': \ :
M:-:' |: /
BO
; : ; : _ j : : :
.^H: . : :
50"::.:
MM ::.: : : -: :
::: ~-.:::: ML :::: LML : L
- • j - . ::i
2 0 ;;i! : : : : ; i :
;;:: ii:: WY- :::: ::':': :"\ ::.
.::: L IM :LLM
o t":l;::;r - iiiiliii-
' !' ; ; ! ; : j ; 1 ; ; ' . j " .'..'
' ! • j ' • i ; i j j : i •j • '! - ! ::.: ; j • • | ! ! !
j .. • . : ... ' ' j • i • . ! : . ! . . . . : . j .
MLMLL^M". j L MM :M'.ML M . : . , . . . : M: " , : . : | : : ML; M M M i - M •• • • • ; : : ! • :
^ • • • : . • • L ^ , ^ . — * - ' " '
j ^ J - - - " ' : ' . - . . • . : :::: ^ j ^ n * " " '.
_ J * . - ' " * : : : . ::•::. y $: ,* 2 :::: ; , . . L ^ ^ ^ ^ p L L , : , , ; , . ; , . ; , ; ; , : . • , . .. - ^ _ _ ^ ^ S * • * MM MM
• . .": M : . . p * " " * " ' ^ :: | ' . ••.'' . '!.: ^:H
• i * * - i i • j • " " " • • - . j <j( '/2 ' !>»•*: :: ' M : . . i j j i ^ . • *» • ." "•
M.(<. ' . ! . • : ! ' ' j .' 1 _j_-.~r" : ": j ' j* •• i -_^—jr"*V j i :
:- * * ' — L " M ' : ' • I • • ' ' " ! | • 1 ' ! '! i ; M .jiiif ;:: j -: j j
•^»: M . ".j..- r - « Uj iS
rr;;::;:i;: :: 'j-plflfnri ..^jii-fir*.. : :• . 1:. . ! :
•MMM** - * - " - ' ! . -L . f :
r 1 ^ • • j . : : ; ; . i
• • • | ' ' • • . . • • . : • • : . : . •
• ' ! L ' i ' !• M • ". ..:
i • .- i : : i ! • ,| • ;
j - : : . ..: , r
. : . . . : " . . . : : ' ' ' . . " . . . : • ' . . ' : ' : ' • . : • • • :
. . : : • . : : . : • . . . : : • . : : . -
LML !
: - - l \ : - -:^-: . - T : \ :
•• - • ••• , ' • • ] :• • • '-]•• • • '••] ]'••• : - — - M M ' M i •]•••••••••••• ••••••••• ' • • • • • • : • , : • : • • • • • : • • • • : • " - " • ' - • " • : : '
: • : : : : : : : • • . . : : : : : : - : : : . : . • : : • • : : : : : . : : : : : : : : : : : :::: MLL ••!• " :
• • : • • : : . , : : : : , : . : : ^ • - .
1 :::|MM|M: |: , |: :::|;M:|ML [ : ; ;| : : : : ; : ; | ; . . : [ :H![LLL| ; ; : | : , ; ;. :|.;.:|:LL;[;L;;|;:.;|:;:: : : : ;| ; : : :| ; LM|ML;[ , | : ; ; , , j ; : ; : | ; . :|
Fig.2.15: Depth-duration-frequeney relation (La Oliva - northern Bardenas).
22
p nm.
I..0
70-
,
•>n
10
0
- • ! - •
-;-
!
...|_
-f-
--4
'
- - -
- j -
._.;_
•
i
!" ! : i
-+--
""h"
- t - :
-4-
-:--
-{-4 -
! •
~f~
--4
.._!_..
_|_.
-4-
4 -
: ' ! : j
; . i
i :
1. '•
:
"4-
• ~ j - -
-4-
•~4 i
i
- I -
- j ^ - 4
-4-
- ~ j -
-~j~ 1
i
._._,__ :
i^1
j
i
1
-4-
- 4
-f--{--
i
. [ . j -!
- T -
( . „ .
- j - - ~ ,
|
4 -
-4-
i
- 4 -
-1- -
-f-
5
........
" ';"-'
....;...
-4—
j i
-f-
i
-4-_..
. . . . ) : -
-4-
j
. ._ ) ._ .
-+•"
i
1
'
—
-J
...j...
j
-+—
..:
|-:;i-:-l i
: l.
.. ! 1 :
':" i
" I .
- I' T
;-
' ! •
j
i
- t -
-
::
._:
...i'.sj
: 'i:-:
•rs
\
<Q
Si
; j
j
i* t*
! V 1
1
j
• • •
; : ; :i ;- ; :
; -i -
'•
ft* :
cl
ril
i '
;
4
| ;
| •
'. '
;
: • 'f
" j - '•'
: .1. . :::.-: : ;: '
: : i : : : [. .:•
1 -
h
4 ... j . .
• i -
! • •
'••'• | ~":
:::-|: DA YS
Fig.2.16: Depth-duration-frequency relation (Ejea - southern Bardenas).
In the northern area the maximum 3-day rainfall with a 5-year return
period would be 76 mm.The corresponding amount in southern Bardenas is 85 mm.
Taking into account only the irrigation season 68.5 and 75 mm respectively
are the amounts to be expected in 3 days for a return period of 5 years
(Fig.2.17).
23
TABLE 2.2 Potential evapotranspiration data
Mean potential evapotranspiration in mm (Thornthwaite method)
1 2 3
H<
3
36 9 . 8 18 10 .1 56 12 .4
an p o t e n t i a l ev
56 2 0 . 5
ap
14 .8 1 5 . 3 17 .4
o t r a n s p
3 2 . 6
3 2 . 6 3 8 . 2 37 . 1
i r a t i o n i n
7 1 . 5
5 1 . 8 7 3 . 2 52 . 4 9 1 . 2 5 3 . 4 8 6 . 5
mm (Penman method)
9 7 . 5 143 .4
110 122 121
170
3 1 3
6
Mean potential evapotranspiration in mm (Turc method)
3 56 2 3 . 8 3 6 . 9 6 4 . 5 9 1 . 0 120 .4 146 .9
136 .2 125 .2 8 9 . 5 5 3 . 2 2 2 . 6 11 .8 7 31 . 0 145 .8 131 .7 9 4 . 2 5 3 . 7 2 4 . 6 13 .8 793 .1 145 .0 135 .5 9 3 . 6 5 4 . 6 2 4 . 6 14 .3 795 .7
194 .6 164 .8 9 5 . 0 5 4 . 3 2 2 . 4 15.1 1082 .3
164 .2 143 .2 100 .8 6 5 . 0 3 7 . 7 2 3 . 6 1018 .0
NOTE: Station J Station 2 Station 3
La Oliva Gallur Zaragoza
12°22' 41°52' 41°39'
- 1°27 - 1°19 - 5 3 '
W W W
342 m 257 m 237 m
The variation of mean daily potential evapotranspiration (E-Thornth-
waite) is shown in Fig.2.19. There is a good agreement between the average
E-values (Thornthwaite) for the south of the area and Zaragoza meteorological
station.
c : : : i ! =:== ==1=:::: :==: ^iJi^WlliMl h=5 H=: Ii:: ::U H=l l==! Hi! 1:1= H== ;i::::::!!::l:!i::iii:: ! : ! ! ! : ; : : £ *? ! ! IS!!;!!:!: :::i :!H I:!:!:::
. . • I ; ! ; ; - " : : ; = ! : ; ; ; ; : - \<$$ oiijim ^ :=-3SLviiiii-ii::ii;::l-:;:i::^SSiiiSH ^ - , _ . „
_ •-.::::: :::v^f\- \.\\^S.A. : ;;;!*&)£:- :•:; ;::: • : : : i ; ^ : ! : ; - ; : : ^ : . :n i | i $ i? : i : : ;!%;:::•:::;:;: ;.•:• _
s . . - •^^:-.jr,Mv\ •i::i;!S^::!;:::;,.-i:;:7f_ ^ ^ : : \ ^ - ^ ..:::;• ; i ^ ^ ^ I T
J ^ . i f : , , ^ ::•;;::: /•::; A : , ^ •tt: #£ .^ -itxA^^tZ. , . J 2 ^ • j^X~71_,l_^^ _^ ' X"'-.^--' -h.-- • •-.•-:-:%^. !
" J F ^ " - j I.: : : : : : . . : . : : *f$=5
J F M A M J J A S O N D Zaragoza ( Penman) La Oliva (Thornthwaite)
Gallur (Thornthwaite)
Fig. 2.19: Mean daily potential evapotranspiration.
Comparing the E-values derived by the Thornthwaite method with those
of the Penman method (Zaragoza) a good agreement is observed during autumn
(September - December), but for the rest of the year Penman values are
greater. This is explained by the absence of wind in autumn.
26
The actual consumptive use of irrigated crops has not been calculated
but in the irrigation season the total deficit of precipitation ranges
between 580 and 700 mm. The peak E-values of 6 and 6.5 mm day occur
in July.
2.4 Natural vegetation
A great part of the area is under crops, so that natural vegetation
is restricted to residual and eroded soils not used for agriculture and to
salt-affected soils.
The natural vegetation cover is of mediterranean semi-arid type and
can be broadly grouped into the following plant associations:
1. In the northern area on the edges and escarpments of mesas roughly
above the 400 m contour line, there is a sparse forest of P-Cnus halepensis
with bushes of Queraus ooooifeva and Pistaoia (Fig.2.20).
Fig.2.20: Sparse forest of Pinus halepensis.
27
2. Below that altitude on residual salt-free soils there is a plant
association of bushes of Juniperus, Thymus vulgaris, Rosmarinus officinalis
and Genista.
3. Halophytes grow on the saline soils of alluvial valleys and fans.
Dominant species are Suaeda brevifolia, Plantago arassifolia and Hordeum
murinwn (Fig.2.21). On the borders of fields and along the roads bushes of
Atriplex halimus, Salsola vermioulata, and Artemisia herba-alba grow where
salt concentration is not too high. In fluvio-colluvial valleys and plains,
where drainage conditions are so poor that a very saline groundwater table
is close to the soil surface, Arthroanemum glauown and Suaeda are found.
Near water courses of saline areas there are isolated trees of Tamarix
afriaana.
Fig.2.21: Halophytes on a saline soil.
28
4. In patches of slight salinity in the northern alluvial plain
there is a transitional subhalophyte association in which Lygeum spartum
is the dominant species (Fig.2.22).
Fig.2. 22: Subhalophyte association with Lygeum spartum.
Along main water courses and roads new plantations of Populus have
been established and round new villages Pinus halepensis is grown in
public parks.
Plant associations of the Ebro Basin were investigated by Braun-Blanquet
and O.de Bo 16s (1957).
29
2.5 Present land use
2.5.1 Irrigated farming
Irrigated farming is influenced by soil conditions. Salt-free soils of
mesas and alluvial valleys are under full irrigation, the main crops being
maize, lucerne, sugar beet, and some horticultural crops. Crop rotations
include winter cereals, mainly barley and wheat, which are irrigated in
spring. Soybean is a new crop which is being introduced in the area.
Seed-beds are prepared by ploughing late in winter, heavy soils
requiring further harrowing and rolling. Summer crops are sown in April
and early May. Harvesting starts in late October, with the harvest of
beet continuing until January.
Lucerne stays on the field for four or five years, depending on the
effective depth of soil. Sowing time varies from late summer to spring.
Sowing in autumn involves a risk of damage by early autumn frosts. On the
other hand in clayey soils dry spring winds can form surface crusts which
may hamper germination.
In the piedmont and colluvial slopes where salinity levels are low,
winter cereals are grown. In spring one or two irrigations are applied
depending on weather conditions. Because land levelling exposes deeper
saline and less pervious layers, most parts of the slopes are unlevelled.
For cereals spring irrigation is done by gravity flow which means an uneven
water distribution. When slopes are irrigated by sprinkler irrigation,
lucerne is included in the crop rotation.
Cropping pattern and intensity on the saline soils of alluvial valleys
and fans depends on their salinity level. Crops include barley, sugar beet,
and lucerne, although lucerne is only grown on successfully leached soils.
Severely salt-affected soils are not cultivated.
Average yields per hectare obtained on some soil units are listed in
Table 2.3.
30
TABLE 2.3 Average yields (kg/ha) for different soil units
Crops
Maize
Lucerne
Sugar beet
Wheat
Barley
Soils of the
mesas
8 000
40 000
30 000
3 500
3 500
non
7
55
40
4
4
Alluvial
saline
000
000
000
500
000
si
soils
ightly saline
-30 000
30 000
3 500
3 500
Soils of the slopes
slightly saline
2 000
3 000
Farm operations are completely mechanized, except for sugar beet grown
on small areas, where much manual work is required in the singling and
harvesting stages. The rise in labour costs is the main reason for not
increasing sugar-beet acreages.
Farmers use mineral fertilizers on a large scale but as animal hus
bandry is uncommon in the area manure applications are limited.
Grain and fodder are sold and sent outside the area. In the villages,
farmers keep some milking cows and pigs for their own use.
Rural industries are mainly confined to maize-grain drying and to
lucerne drying, but some vegetable processing and canning has commenced.
The size of newly established holdings was about 8 hectares at the
start of the settlement (1955). Nowadays, new farms average 20 hectares
of irrigated land. In addition, the settlers receive some non-irrigated
arable land.
2.5.2 Dry farming
On the higher lands not under command of irrigation barley is grown.
Seed-bed preparation starts in early autumn. The most common date of sowing
is late October, but depends very much on the rainfall distribution. New
wheat varieties less sensitive to spring frosts are now being introduced
in the area and sowing time is delayed until January. Harvesting starts
late in June for barley and in July for wheat.
Residual soil areas unfit for cultivation are used as rough pasture
for sheep.
31
and sub- la te ra l canals are l i s t ed in Table 3.2.
TABLE 3.2 Secondary canals of Bardenas-1 area (From F.de los Rios, 1966.)
Canal Total
length (km) Flow
m3/sec Origin
Pardina
Navarra
Cinco Villas
Cascajos
Saso
7
36
51
23
11
29
8.7
17.8
7.8
7.5
Bardenas
Pardina
Pardina
Bardenas
Bardenas
At present the Bardenas canal debouches into the Arba river, but the
aqueduct of Biota over the Arba has now been completed and the second part
of the canal is under construction. When it is completed, it will irrigate
the second part of the Bardenas area (approximately 30 000 hectares more).
Quality of the irrigation water
The Bardenas canal water is of good quality for irrigation as its
electrical conductivity (EC.) varies from 250 to 750 micromhos/cm.
The small amount of leaching water needed to maintain a low salinity
level in the irrigated soils is met by the normal application losses
associated with surface irrigation.
There is no marked annual variation in the salt content of the irriga
tion water, though salinity increases slightly towards the end of the
irrigation season. Table 3.3 gives the mean composition of the irrigation
water in 1972/73.
34
TABLE 3.3 Bardenas canal irrigation water analysis
Total
Cl~
ro3=
C03H~
S0.=
4
Ca +
+ Na
K+
++ Ca
Boron
EC-25°
pH
SAR
RSC
Class:
salt content
meq/1
meq/1
meq/1
meq/1
Mg meq/1
meq/1
meq/1
meq/1
meq/1
meq/1
C micromhos/cm
meq/1
fication
Winter
6.8
0.5
0.1
2.2
0.3
3.0
0.7
-
2.8
-
280
7.8
0.6
0
Vs.
Spring
6.5
0.3
0.1
2.9
0.2
3.2
0,2
-
2.4
-
305
7.5
0.2
0
VS1
Summer
6.9
0.5
0.2
2.4
0.4
2.9
0.4
-
2.3
-
295
7.7
0.3
0
Vs,
Autumn
7.5
1.4
0.1
1.6
0.8
2.2
1.4
0.1
1.5
-
345
7.7
1.3
0
Vs.
The sodium adsorption ratio of the irrigation water is in the lowest
range (S.), and the residual sodium carbonate (RSC) value is zero, so that
there is no danger of alkalinization.
3.1.2 The main drainage system
The area studied can be divided into five hydrological units, each
drained by a main water course (Fig.3.2).
Unit A drains toward the Castiliscar stream, which joins the Aragon
river at Carcastillo. The unit receives water from rainfall, irrigation,
and run-off from the surrounding uplands. Former intermittent streams
have been deepened and are now used as main drains.
35
,/ (Tit)
• major streams irrigation conveyance canals watersheds
Fig.3.2: Schematic map of the main irrigation and drainage systems.
36
Unit B drains directly toward the Aragon river. Ephemeral water
courses flowing through the Aragon alluvial valley are used as main drains.
Unit C, apart from receiving rainfall and irrigation water, also re
ceives seepage water from the adjacent western uplands. Intermittent streams
flowing down join the Valarena stream, which receives supplies from the
adjacent mesa as well. The Valarena debouches into the Riguel river in the
south of the area.
Unit D drains toward the Riguel river, which receives water from
irrigation and rainfall on the Riguel alluvial plain and from part of
the two adjacent mesas. Water courses in the erosion valleys of the mesas
have been deepened and are now used as main drains.
Unit E drains directly toward the Arba river and comprises the eastern
edge of the eastern mesa and the southern part of the studied area.
Generally the salinity of the water in the main drains is low. In fact
in 197A/75 the EC of the Aragon water at the north-western end of the area
was never higher than 0". A mmhos/cm. In the same period the Arba water at
the south of the area had a variable salinity between 0.9 and 4.5 mmhos/cm.
The latter value is more representative of the salinity of the drainage
water, as the Bardenas canal tail escape flows into the Arba river.
The value of 4.5 mmhos/cm was found when the canal ran dry (February).
Likewise the salinity of both the Castiliscar stream and the Riguel varied
between 2 and 5 mmhos/cm.
As the salinity of the groundwater is very high in some geomorpholo-
gical units (see Tab.3.4), the low EC-values in the main drains indicated
that the saline soils were being leached at a very slow rate. The greater
part of the area's drainage water comes from the tail escapes of the irri
gation system, from non-saline groundwater, and from run-off. Very little
is derived from the saline areas, which are not cultivated and therefore
not irrigated.
37
3.2 The groundwater table and drainage conditions
The depth and hydraulic gradient of the groundwater table and the
salinity of the water depend on the situation of each geomorphological
unit and its relation to adjacent units.
The drainage conditions in each unit depend on the depth of the least
permeable layer, the hydraulic conductivity of the permeable layer and its
drainable pore space, and the recharge of water received. The recharge may
be supplied by rainfall, by net percolation from irrigation, or by seepage
from surrounding high-lying areas.
In the project area some units of a characteristic hydrology can be
defined in relation to the major geomorphological units.
3.2.1 Rough mountainous land
In the uplands that surround the irrigated area, the effective
rainfall is very low because the unweathered mudstone and siltstone are
impervious and so prevent infiltration. Most of the rainfall becomes surface
run-off, which causes erosion of the soft sediments. Ephemeral streams
catch this run-off and transport the water into adjacent low-lying areas.
3.2.2 Aragon alluvial valley
No water table has been detected in the upper 3 m of the fluvial
terraces of the Aragon valley. The whole unit drains freely toward the
Aragon river.
The transmissivity of the coarse alluvium (k > 2 m/day) is so high that
no drainage problems have been created by the irrigation of the area. The
impervious layer is formed by the underlying mudstone, which is always
deeper than 3 m.
There is local seepage from the terrace escarpments but no deep seep
age from higher to lower terraces because the underlying mudstone is imper
vious.
38
Although no groundwater samples were taken, its salinity is assumed
to be very low because it is recharged only by rainfall and by irrigation
percolation losses.
3.2.3 Rough eroded plain
In this major geomorphological unit, drainage conditions and water
table depths differ in each minor unit.
The fluvio-colluvial valleys receive water from rainfall, irrigation,
and seepage from the surrounding residual hills and slopes. The depth of
the impervious layer varies between 1 and 2 m and the transmissivity of the
permeable layer is low (K = 0.2 to 0.5 m/day, D = 1 to 2 m ) . The hydraulic
gradient is low as the slope of the unit is less than 1 per cent. The
lower-lying valleys thus have a shallow groundwater table and poor drainage
conditions.
In the eroded and colluvial slopes there is no indication of a water
table within 2 or 3 m of the surface. At present, however, the eroded plain
is not under full irrigation, the slopes being only locally irrigated in
spring. Any increase in water use could cause the groundwater tables in
the lower valleys to rise. If that were to happen, a cut-off drainage
system would be sufficient to control the lateral seepage, as the pervious
layer in the valleys is restricted to the mixed alluvium-colluvium and the
impervious layer is not deeper than 2 m.
The salinity of the phreatic water of the fluvio-colluvial valleys
is very high, sometimes reaching EC-values of about 50 mmhos/cm (Table
3.4).
As the lower areas do not drain freely toward the main drains, a sub
surface drainage system is necessary to allow the soils to be leached and
to prevent a rise of the saline water table.
39
TABLE 3.4 Characteristic groundwater analysis
G e o m o r p h o l o g i c
Cl
co3=
CO 3 H"
S °4 =
Ca + +
Na+
K+
Ca + +
Boron
meq/1
meq/1
meq/1
meq/1
+ Mg meq/1
meq/1
meq/1
meq/1
meq/1
EC-25 C mmhos/cm
pH
SAR
RSC meq/1
Classification
Piedmont Riguel Erosion Fluvio- Saline Saline slopes alluv. valleys colluvial alluvial alluvial
plain valleys valleys fans
894 . 4
2 .4
79 .7
228 .6
751 .1
0.1
72.8
62.9
7.8
70.2
2.8
0.4
3.9
4.6
5.8
6.6
-
4.5
1.2
7.9
3.9
Vs.
11.7
0.1
8.7
7.5
9.5
19.5
-
5.6
2.7
7.6
8.9
VS3
23.9
0.3
3.9
14.3
20.0
23.3
0.1
10.9
4.2
7.5
7.4
VS3
38.5
0.4
4.3
24.2
11.8
56.8
0.1
4.9
6.4
8.0
23.4
VS4
348.0
0.6
3.0
133.1
102.2
380.0
0.1
36.0
48.0
7.7
53.2
VS4
342.2
-
11.9
55.0
107.0
304.1
-
51.1
32.8
7.2
41 .6
VS4
3.2.4 Alluvial fans
In this unit with the present low intensity of irrigation, there is
no indication of a groundwater table within 3 m of the surface. Deep
borings showed water levels between 5 and 6 m.
3.2.5 Alluvial plain
Drainage conditions and water table depths in this unit are similar
to those of the fluvio-colluvial valleys. The hydraulic gradient is even
lower than that in the valleys as the slope is smaller.
A deepening of the drainage outlet and an initial drainage system
of open ditches have allowed some control of the water table with the
40
present low rate of irrigation. When the area is more intensively irrigated
a more efficient drainage system will be required.
3.2.6 G y p s u m ridges and valleys
The gypsum valleys receive water from the adjacent parallel ridges
and from the higher lands to the east. They are also recharged by leakage
water from the tail escape of the Navarra canal. As the area is not
intensively irrigated losses of irrigation water are negligible.
At present there is a drainage system which consists of a main open
drain running along the valley and several lateral ditches. The system is
sufficient to control the water table which in the lowest parts of the
valleys is still more than 1 m below the surface. When the area is under
full irrigation the present drainage system will need to be improved.
3.2.7 Mesas
The mesas receive water only from rainfall and irrigation as they are
at the highest level of the irrigated area and are cut off from the higher
surrounding uplands. The permeable layer of coarse alluvium bears a perched
water table above the underlying mudstone which acts as an impervious layer.
The transmissivity of the coarse alluvium is high enough (K > 2 m/day,
D = 1 to 3 m) to provide adequate natural drainage.The unit drains freely
toward the mesa edges and finally to the main drains of the erosion valleys.
For this reason there is no permanent phreatic aquifer but an epheme
ral perched water table. Its depth varies locally depending on the water
management. There is also a seasonal variation due to the unit being under
full irrigation. The water table is at its lowest in the non-irrigation
season, but even during the irrigation season it is never high enough to
affect crop growth.
As the water table is recharged only by rainfall and irrigation, its
salinity is low (EC < 2 mmhos/cm). Diffusion of salts from the underlying
saline mudstone, where it exists, is almost negligible.
41
3.2.8 Eroded mesas
In small depressions of the mesas where the coarse alluvium is shallow,
and in the erosion valleys of the mesas where the alluvium has been removed,
permanent water tables can exist, if they are fed by lateral seepage from
the higher mesas.
Where an open drain leading from such depressions through the erosion
valley has been constructed, the local drainage problem has generally been
solved.
In the depressions the salinity of the water may rise to EC-values
of about 9 mmhos/cm.
There is no deep seepage from the mesas to the low-lying alluvial
valleys because the mudstone underlying the coarse alluvium is totally
impervious. In the mesa escarpments, however, there is lateral water flow
which creates local surface seepage. Cut-off drainage ditches intercepting
this local seepage prevent the waterlogging of lower lands.
In the piedmont slopes between the mesas and the valleys, the imper
vious layer is formed by the mudstone or siltstone underlying the local
colluvium. This impervious layer tends to prevent deep percolation, thus
creating surface run-off in the upper shallow part of the slope and shallow
interflow through the lower part.
If the natural gradient of the impervious layer has not been disturbed
by land levelling, the slope drains freely towards the lower unit. Piedmont
slopes therefore do not have a permanently high water table and epheme
ral levels are extremely variable, although generally deeper than 2 m.
The salinity of the water varies between 2 and 8 mmhos/cm.
3.2.9 Riguel alluvial plain
The permeable alluvium (K = 0.3 to 0.5 m/day) extends deeper than 3 m.
For this reason natural drainage is generally sufficient to prevent water
logging and salinization.
42
The whole unit drains freely into the Riguel river. The water table
depth is therefore determined by the water level in the river and locally
by the water management of individual farmers.
The salinity of the water ranges between 1 and 5 mmhos/cm (Table 3.4).
In the lower part of the plain, where seepage from the higher units rechar
ges the water table, the salinity of the water increases up to 10 mmhos/cm
due to a high sodium chloride content.
3.2.10 'Badlands', alluvial valleys and fans
The residual ridges and eroded slopes have shallow impervious sub-hori
zons which prevent infiltration. Thus most of the rainfall is removed as
surface run-off and recharges the lower-lying alluvial valleys and fans.
The soils of the alluvial formations have surface horizons whose
hydraulic conductivity varies from low to moderate (K = 0.1 to 0.5 m/day).
Very fine stratification frequently occurs in the subsurface layers.
Where it exists, hydraulic conductivity for vertical flow (percolation
rate almost zero) differs from lateral permeability (K = 0.2 m/day). Occa
sionally the sub-surface layers are compact and impervious. Below these
impervious layers a gravel and coarse sand alluvium often occurs, which
in turn is underlain by totally impervious mudstone.
Rainfall or irrigation water remains on the soil surface until it
evaporates and sometimes a temporary perched water table is formed near
the soil surface. In this way surface run-off is common even in levelled
soils, thus creating a severe erosion problem.
The deeper coarse alluvium (between 3 and 5 m) is always saturated
with water and forms a confined aquifer between the impervious mudstone and
the confining stratified or compact layers.
Though a specific geohydrological survey of the confined aquifers
was not made, they seem discontinuous because the gravel layers have a
variable thickness and were not found in some deep bores.
43
In addition to the recharge from rainfall and irrigation, further
water is supplied by leakage from the Cinco Villas canal and from uncon
trolled seepage from higher irrigated areas.
The salinity levels of both the perched and confined water are both
high and variable (EC = 10 to 75 mmhos/cm). High electrical conductivity
values are caused by high sodium chloride contents (Table 3.4).
With the exception of some low-lying areas, there is no indication
of a true groundwater table within several metres of the surface. It was
therefore useless to attempt to draw an isohypses map of a phreatic aquifer.
44
4. Soils and soil conditions
4.1 General characteristics
4.1.1 Soil-forming processes
Most soils of the Bardenas area show no distinct characteristics of
soil development. Except in the ancient alluvium of the mesas, soil-forming
processes have had little time to act upon the very recent parent materials
of alluvial and colluvial origin. Moreover, the climate being characterized
by a marked aridity, an excess of water rarely occurs owing to the low
amount of precipitation and its even distribution. Therefore, except for
the soils of the mesas, there is little soil profile development. The soils
of the mesas, formed over a much longer period, have diagnostic cambic and
calcic, or petrocalcic horizons. The other soils, apart from an ochric epi-
pedon, show the beginnings of argillic, calcic, gypsic and even salic hori
zons but only in a very incipient stage.
Two soil-forming factors are relevant in understanding the origin and
extent of the salt-affected soils of the area. These factors are parent
material and the relative position of each soil association within the
landscape.
4.1.2 Parent materials
Parent materials can be grouped into three classes. First is the ancient
coarse alluvium of the mesas, which consists of well-rounded stones, gravel,
and coarse sand. Above this layer a reddish loamy generally stoneless layer
occurs. Both the coarse and the reddish layers are rich in calcium carbonate
and are salt-free.
45
Second are the parent materials of alluvial origin brought by the two
main water courses flowing through the area - the Aragon and Riguel rivers -
whose catchment areas are respectively outside and just at the edge of the
Miocene-Oligocene sedimentation basin. The parent materials of the Aragon
valley soils are salt-free coarse alluvium in the medium and higher terraces
and loamy sands in the lower terrace. In the Riguel valley a fine alluvium
is found, which consists of clay loam and is salt-free in the upper part of
the valley. Both clay content and salinity increase towards the lower end
of the valley. A narrow sandy levee not sufficiently wide to be represented
on the map can be distinguished.
The third class of parent materials are those derived from the saline
mudstone and siltstone, which are characterized by a very high percentage
of silt. In the alluvial valleys and fans a fine alluvium has been deposited
and in the lower parts of the slopes a fine colluvium. Both occur mixed in
the fluvio-colluvial valleys. In the upper part of the piedmont slopes the
parent material is mainly weathered mudstone. As these rocks are widespread
throughout the area, soil texture varies within a narrow range, silt loam
and loam being the most common textures. Salinity, though always present
in these parent materials, is highly variable. In general the amounts of
salts are highest in the low-lying alluvial formations (Chaps.2 and 3).
High calcium carbonate content (>40%) is a common property of the soils.
In non-saline soils pH varies between 7.5 and 8. In soils with higher sodium
content higher pH-values are common (8 to 8.5).
4.1.3 Topography
The soils of the mesas, fluvial terraces, and alluvial valleys and
plains are flat. The eroded plain has an undulating topography with gentle
slopes and flat low-lying fluvio-colluvial valleys. Stony ridges and mesa
escarpments have a pronounced slope.
Topography has had a marked effect on the redistribution of salinity.
The colluvial and piedmont slopes have lost salts which have accumulated in
the lower flat areas (Chap.3).
46
4.1.4 Land and water management
Where the soils of the slopes have been levelled, the original soil
profile has been truncated. Sometimes less pervious and more saline layers,
previously deep, form the present surface soil. The flat soils of the mesas
and alluvial valleys required only a slight smoothing of the land and the
present soil represents only a slight re-arrangement of the original pro
file.
Irrigation has also played a role in the redistribution of salts. The
percolation losses inherent in surface irrigation have leached the salts
from the more permeable soils and carried them to other soils, particularly
to soils deficient in natural drainage.
4.2 Mapping units
During the reconnaissance survey which formed the basis upon which the
soil map was prepared, the physiographic approach was applied. The objective
of the reconnaissance survey was to delineate the saline and potentially
saline areas for their subsequent reclamation under irrigation. The mapping
units are therefore based on those conditions that were relevant to the
specific practical agricultural purpose of the survey, namely the salinity
hazard and the land reclamation possibilities.
The mapping units thus represent landscape units based on the geo-
morphological features of land forms, slope, elevation and relative si
tuation, parent materials, and drainage conditions. As has been described
in the preceding chapters, all these features contribute to soil salinity.
The soil map prepared in this way is shown in Figs.4.1 and 4.2. The
main mapping unit is a landscape or geomorphological unit. Some subdivisions
were made on the basis of soil salinity level and drainage conditions.
Erosion hazard also had to be added because it is a condition that radically
affects the land suitability for irrigation.
47
Each mapping unit is equivalent to a broad soil association but with a
similar salinity range and similar drainage conditions.
Some miscellaneous land types are also included. They indicate land
with little or no soils, which serve no purpose for irrigated agriculture,
and whose soil conditions were therefore not studied.
To standardize the various investigations and to ease correlation with
soils of other areas, representative soil profiles were morphometrically
classified. Descriptions of soil profiles are in conformity with the Soil
Survey Manual (1951). Most of the soil analyses were done in line with
those of Agric. Handbook No.60 of the USDA (1955). Soil classification was
based on Soil Taxonomy, Agric. Handbook No.936, USDA (1975).
Four main zones, each of them related to a drainage basin, can be
distinguished on the soil map (Chap.3).
The northern "Rough Eroded Plain" has a complicated soil pattern
(cross-section in Fig.2.8). It comprises residual siltstone outcrops in the
highest position and saline alluvial soils in the low-lying fluvio-colluvial
valleys and plains. The transition between the two mapping units is formed
by soils of eroded slopes in the upper part of the hills and soils of collu-
vial slopes in the middle and lower parts of the slope.
The north-western part of the area is dominated by the "Aragon Alluvial
Valley" (see cross-section in Fig.2.7) where soils of fluvial terraces
are found. In the south this zone is bordered by a high mesa situated out
side the irrigation scheme. Between the alluvial soils of the valley and
the higher soils of the mesa, transitional soils occur on piedmont slopes.
The same soil pattern dominates the south-eastern zone of the area as
well (Fig.2.10). The soils of the mesas occupy the highest position and
the soils of the Riguel alluvial plain the lowest. The soils of the piedmont
slopes form a transition between the two.
The south-western major zone of "Piedmont slopes" has a highly irregular
48
soil pattern (Fig.2.12). It is composed of residual soils of the stony
ridges and miscellaneous land in the highest position, and saline alluvial
soils in the low-lying alluvial valleys and fans. Eroded slopes are situ
ated between the residual soils and the depositional ones.
The mapping units and their main features for irrigated agriculture
are described in the following paragraphs.
Fig.4.1: Physiographic soil map of the Bardenas area. Sheet I - Aragdn basin. (Base map drawn from aerial photographs on an approximate scale of 1 : 31 000).
For legend, see p.50.
49
4.2.1 Mesas
Mesa soils (Mapping Unit 1.1) are the more developed soils of the
area because the ancient coarse alluvium from which they derived had in
filtration and percolation rates high enough to allow an illuviation of
calcium carbonate and even clay during former periods of wetter weather.
Calcic (Profile 2) and petrocalcic (Profile 1) horizons were thus formed.
Part of the original surface soil has been removed by erosion so that
the soils are shallow with a limited rooting depth. On lower-level mesas
however, a deep reddish fine soil overlies the semi-consolidated alluvium.
In some small depressions the coarse alluvium is not found and the soils
lie immediately on the impervious mudstone.
This mapping unit is a broad soil association in which effective depth
of cultivable soil and soil texture vary (Fig.A.3).
Fig.4. 3: Landscape of a soil of the mesas.
52
As the purpose of the survey was to map saline areas and as this soil
association is in general free of salinity, phases of slightly different
soils were not defined, except in an isolated phase (1.2) for remnants of
eroded mesas which are out of the irrigation command.
The general soil properties of the mesa soils are:
a) Depth above the petrocalcic horizon varies between 30 and 80 cm.
Root proliferation is well developed in the surface soil and even the petro
calcic horizon is occasionally penetrated.
b) Textures vary from sandy loam to silt loam, although soils may
become more sandy with depth.
c) In the shallowest soils the degree of stoniness is high, but in
general the surface soil is free of stones and gravel.
d) Structure is weakly to moderately developed although the surface
soil generally has a strong crumb structure.
e) Infiltration and percolation rates are more than adequate to
allow water to pass through the soil profile. The transmissivity value is
sufficient to permit adequate drainage (Chap.3). Total moisture-holding
capacity varies with the soil textures and effective depths of soil.
f) Salinity is uncommon but slight salinization can occur in depres
sions where the coarse alluvium is only shallow over underlying impervious
mudstone (Chap.5).
g) There is a rapid increase in calcium carbonate content with depth,
being highest in the calcic and petrocalcic horizons.
h) Soil colour above the calcic and petrocalcic horizons is normally
7.5 YR 4/4.
Where the soils of the mesas are moderately deep (Fig.4.4), prosperous
irrigated agriculture has been established. Maize, lucerne, sugar-beet,
barley, and tomato are the main crops.
Effective depth of cultivable soil and moisture-holding capacity are
the main soil features on which land suitability for irrigation depends.
53
Their general properties are:
a) Depth is generally greater than 2 m, but root proliferation is
limited to 50 cm.
b) Textures vary down the profile and signs of stratification are
evident, as could be expected in young alluvial soils. The surface textures
are mainly clay loam. Stones are rare.
c) Structure development is weak and often non-existent, particularly
in the subsoil.
d) Soil moisture-holding capacity is the highest of all the soils of
the studied area. Infiltration and percolation rates, though low, are suf
ficient to allow surface irrigation. In general no water table is found in
the upper 3 m (Chap.3).
e) The surface soil is free of salts but below 2 m soil salinity
increases to an overall value slightly higher than 4 mmhos.
Profile 5 represents a characteristic soil of the middle valley. The
clay content increases in the lower part of the plain (Profile 7). The narrow
natural levee of the river consists of very recent sandy soils (Profile 6,
Fig.4.7).
In the lower valley soil salinity varies in accordance with the amount
of leaching achieved under normal irrigation. It has been separated as a
general saline phase, however, because its overall degree of salinization
is higher than in the upper and middle valley (Mapping unit 3.6). A halo-
phyte vegetation, which does not occur in the non-saline phase (Fig.4.8),
is found in this unit.
With the present drainage system of open ditches desalinization is pos
sible as the percolation rate is sufficient to allow leaching (Chap.5).
Together with the deeper non-saline soils of the mesas, the non-saline
soils of the Riguel river plain are the best soils in the area. They are
under irrigated agriculture, the main crops being lucerne, wheat, and
barley.
58
Fig.4.7: Profile 6.
4.2.5 High and middle Aragon terraces
On both the high and middle Aragon terraces (Mapping Units 3.3 and 3.A)
similar soils are found. They derive from a similar parent material which
consists of a coarse alluvium deposited by the Aragon river (Fig.4.9). Above
this coarse substratum a finer layer occurs (Profile 8).
59
Fig. 4. 8: Profile 7. Fig. 4.9: Coarse alluvium of the middle Aragon river terrace.
These loamy alluvial soils show a moderate profile development. Structu
res are better developed than in the soils of the lower terrace and overall
there is slightly more clay present. Incipient clay skins are frequent in
the subsurface horizon. Calcium is present as mycelia but is insufficient to
allow a calcic horizon to be diagnosed. The underlying coarse layer contains
such high levels of calcium carbonate that it is impossible to state which
is depositional and which is intrinsic.
60
In general the profile development of these soils (Fig.4.10) is less
intense than that of the soils of the mesas which are the oldest soils of
the area.
Fig. 4.10: Profile of a soil of the highest'Aragdn river terraae.
Their general properties are:
a) A minimum of 50 cm depth of cultivable soil, limited by a conti
nuous layer of semi-cemented rounded stones. The possibility of root proli
feration into this substratum is only slight.
b) The most common texture of the surface soil is loam, with a slight
increase of clay content in the subsurface horizon. Surface stoniness is
frequent and the coarse fragment percentage increases with depth.
c) There is good structure development in the cultivated surface soil.
Below 25 to 30 cm structure is weakly blocky down to the conglomerate sub
stratum.
61
d) Internal drainage is extremely good and no water table is found.
The infiltration rate is high, and the moisture-holding capacity moderately
low because of the lack of deep fine-textured soil.
e) As the parent material is non-saline and drainage conditions are
excellent, the soils are free of salinity.
The main restrictions for irrigated agriculture are depth of cultivable
soil and surface stoniness, which prevent the cultivation of sugar beet and
lucerne. Cereal crops, maize, and soya bean are cultivated successfully, but
need frequent small irrigation applications.
4.2.6 Low Aragon terrace
The soils of the low Aragon terrace (Mapping Unit 3.2) represent only
a small part of the total area. They are deep sandy loam alluvial soils with
no profile development (Fig.4.11). Stratification is evident at moderate
depth, with gravel and coarse sand layers interbedded between the sandy loam
layers (Profile 9).
The general properties are:
a) Depth of cultivable soil, though variable, is greater than 50 cm.
It is limited by the cobbles and gravel layers which appear at varying
depths. The coarse alluvium is completely loose, as might be expected with
an immature soil.
b) Textures vary from loam to sandy loam. Surface stoniness is rare.
c) Structure is almost non-existent, though slight structure develop
ment appears in the cultivated surface soil.
Soils are excessively drained and no water tables were found in the
studied profiles. The main restrictions for irrigation are the high infiltra
tion rate and the low moisture-holding capacity. Percolation losses under
surface irrigation are therefore high.
62
The soils are non-saline and a permanent irrigated agriculture is main
tained, maize being the most common crop.
Fig.4.11: Profile 10.
63
Land use for irrigated agriculture depends mainly on the effective
depth of cultivable soil and the relative size of non-eroded areas.
4.2.9 Colluvial slopes
The soils of the colluvial slopes (Mapping Unit 2.7) are brown soils
with only an incipient profile development. They derive from a mixture of
colluvium and material from the underlying Miocene-Oligocene sediments.
The only profile development is a cambic horizon underlying an ochric
epipedon, and the movement of calcium carbonate and soluble salts down the
profile. The underlying substratum is generally siltstone, often interbedded
with mudstone. Both are saline sediments with very low permeability (Profile
12).
The general properties are:
a) Average depth of cultivable soil is variable but always more than
50 cm above the slowly permeable substrata. Depth over decomposing mudstone
and siltstone is variable depending on slope and situation, but where erosive
forces are less the depth can be more than 2 m.
b) Textures vary from sandy loam to silty clay loam. The decomposition
of the siltstone produces fine structureless sandy loam textures and the
mudstone silty clay loam textures with partial structure development. There
is very little gravel.
c) Soil structure is not well developed. It is moderately fine granu
lar in the surface soil and subangular blocky in the subsoil. The underlying
substrata are structureless.
d) Soil salinity is variable. Some of the studied profiles were free
of salt and in others salt content increased in the deeper layers. Never
theless there is an overall slight salinity in the unit.
Though most of these soils are under irrigation command, dry farming
is practised. As deeper layers are less pervious and more saline than the
66
surface soil, land levelling is not a suitable practice and irrigation
should be by sprinkler.
A moderate excess of irrigation water over the consumptive use require
ment should be sufficient to leach the salt to deeper layers. Only small
amounts of leaching water should be applied as otherwise saline water would
seep to adjacent low-lying areas and an interceptor drainage system would
be required to cut off the flow.
An eroded phase (Mapping Unit 4.2) has been separated where erosion
processes predominate over the colluvial ones.
4.2.10 Fluvio-colluvial valleys, imperfectly drained and saline-alkali phase
These soils (Mapping Unit 3.8) are light yellowish brown alluvial-col-
luvial soils formed in the bottoms of flat valleys scoured out in the
Miocene-Oligocene sediments. They have developed from a mixed parent material
of fine alluvium from the erosion valleys (2.5), fine colluvium from the
surrounding slopes and hills, and locally decomposed sediments (Fig.4.13).
They show a moderate profile development (Profile 13).
The soils are saturated with water because they suffer from impeded
drainage and are situated in areas receiving seepage from the surrounding
slopes and hills. They are saline and many are alkali as well, but because
of their immaturity no true salic or natric horizon has developed sufficient
ly to become diagnostic. There is a downward movement of calcium carbonate,
and gypsum crystals have accumulated at an average depth of 1 m.
The general properties are:
a) Average effective soil depth is 75 cm. Deeper layers are too com
pact for root proliferation.
b) Their textures are finer than other surrounding soils and are
generally silty clay loam, but clay layers are occasionally found. In the
67
subsoil very fine sandy layers sometimes occur as a result of the weather
ing of underlying siltstone.
Fig.4.13: Profile 14.
c) Structure is moderately good. In the surface soil, structure is
crumby and in the subsurface blocky and prismatic. Below 75 cm there is no
structure development. There are indications of clay movement down the pro
file and clay and silt skins on prism sides and pores are evident (Fig.4.14).
Lime mycelia and gypsum crystals are also common.
d) Salinity is severe because the valleys receive constant saline
seepage from higher levels (Chap.3). The salt content is variable in the
surface soil but increases with depth where it is very high (Chap.5).
68
Fig.4.14: Detail of profile 14.
As these soils are severely saline, few of them are cultivated and
only halophytic vegetation is found. The fact that the salt content in
creases with depth means that infiltration and percolation are probably
adequate for leaching. Thus, if the drainage system were improved and the
amount of leaching water increased by irrigation, these soils could most
likely be reclaimed.
69
4.2.11 Alluvial plains, imperfectly drained and saline-alkali phase
As the alluvial plains (Mapping Unit 3.10) have formed in the flat
areas where fluvio-colluvial valleys coalesce, their soils are very similar
to those of the fluvio-colluvial valleys. The upper 1.5 m is saturated with
water for some period of the year. The characteristics associated with water
logging are insufficient to allow any diagnostic horizons other than a
cambic horizon to be defined. These soils are saline and sodic but again
their immaturity means that no true salic or natric horizon has developed
sufficiently to become diagnostic.
The general properties are:
a) Effective depth of cultivable soil is rarely more than 75 cm
and root proliferation is mainly restricted to the upper 40 cm.
b) Their textures are finer than the soils of the surrounding slopes,
varying generally from silt loam to clay, with silty clay loam as the most
frequent texture.
c) Structure is moderately good with indications of clay movement
down the profile. Lime mycelia and salt crystals with occasional gypsum
are common.
d) These soils have been almost completely drained with an open
drainage system. Below 1 m, however, olive mottling, with the particular
shininess associated with gleying, is found.
Profile 1A is an extreme example of the salinity and alkalinity. Other
wise it is very typical as it is situated in one of the very few areas in
the central swampy zone which has not yet been cultivated.
Some areas of these soils have never been cultivated or at least not
for many years. The drained areas are under irrigation, with sugar beet and
barley as the most common crops. There is an overall high level of salinity.
For the reclamation of these soils, the same considerations are valid as for
the reclamation of the saline soils of the fluvio-colluvial valleys.
70
4.2.12 Alluvial valleys, saline-alkali phase
These young soils (Mapping Unit 3.11) show only an ochric epipedon.
Below the levelled surface soil, they are highly stratified.
The lack of profile development in these yellowish brown alluvial
soils is partly due to their characteristic lamellae sedimentation pattern
which impedes the percolation of water. The infiltration rate is very low,
being almost negligible if the soil is unploughed. Consequently rain water
does not percolate to deeper layers and almost the total amount of rainfall
becomes surface runoff or remains on the surface until it evaporates. There
is no profile development and even salt movement is very slight in the
layers below the surface. *
Profile 15 is a representative soil of this mapping unit, the general
properties of which are:
a) Depth of cultivable soil is equal to the depth of soil affected
by the levelling and subsoiling operations, namely about 50 cm. In the stu
died soil profiles of non-cultivated land.no root proliferation was detected
in the stratified substratum. In fact when the roots of the halophytes reach
the stratified layers, they grow in a horizontal direction.
b) • Texture varies between silt loam and silty clay loam but occasio
nally very fine sandy loam layers are found. The soils are free of gravel,
at least in the upper 1.5 m.
c) The surface soil shows a weak structure development of the sub-
angular blocky type. If the soils remain uncultivated the upper 3 to 5 cm
shows a very fine platy structure with a high salt concentration. Salt puffs
are common on the surface. A strong very fine or fine stratification is ob
served in the subsurface layers (Fig.4.15). At greater depth a gravel and
coarse sand layer saturated with water is generally present (Chap.3).
These soils have a poor structure stability and their erosion hazard
is therefore high, even in flat levelled land.
71
?&&$-
Fig.4.15 A: Soil profile of a saline soil of an alluvial valley.
Fig.4.15 B: Stratified layer of a soil of the saline alluvial valley.
72
d) Pores are common between the lamellae but are vesicular and dis
continuous. Hydraulic conductivity measured by the inversed auger hole
method varied within a restricted low range.
e) The soils are severely affected by salinity and are to a similar
degree sodic (Chap.5).
Of all the soils in the area, these soils are the most unfavorable for
irrigated agriculture (Fig.4.16). Their high salinity severely restricts
crop growth and their lack of internal drainage makes water management
difficult.
An artificial drainage system combined with deep subsoiling is required
to reclaim these soils. If their infiltration and percolation rates were to
be increased with soil management, the surface soil could become non-saline.
Their intrinsic fertility, however, is so poor that an intense fertilization
programme would be needed to enable permanent agriculture under irrigation.
Fig.4.16: Landscape of a soil of the saline alluvial valleys.
Fig. 4.17: Profile 18.
73
Secondly, there are no stratified layers in the soil profile and con
sequently the percolation rate is higher;
Thirdly, these soils have had more time to lose any intrinsic salinity
of the soil parent material.
Profile development is obvious only in a slight downward movement of
calcium carbonate and no diagnostic horizons other than an ochric epipedon
are found (Profile 17).
The general soil conditions are as follows:
a) Although a truly impenetrable layer occurs below 1 m, the effective
depth of cultivable soil is no more than 50 cm because the subsurface ho
rizons are so compact.
b) Textures vary between clay loam and silty clay loam and there is
no surface stoniness.
c) Only a weak structure development in the form of fine subangular
blocks can be seen.
d) As these soils have a shallow impervious layer and the permeability
of the surface soil is moderately low, natural drainage is moderate. Overall
there is a slight level of salinity though some of the soils included are
non-saline (Profile 17).
As the salt content of these soils is lower than that of the soils of
the southern alluvial fans, their suitability for irrigated agriculture is
higher. At present, however, they are not under full irrigation.
An eroded phase (2.6) has been separated from this unit where the ero
sion hazard is so high as to prevent irrigated agriculture.
76
4.2.15 Other minor mapping units
Some minor mapping units with no agricultural significance have been
separated on the soil map. These units are not capable of maintaining perma
nent irrigated agriculture and are generally situated out of the irrigation
command. Their land use possibilities are dry farming in some instances and
more usually sheep grazing.
Such mapping units are the rough mountainous land at the northern border
of the studied area (1.3), the siltstone outcrops in the rough eroded plain
(1.4; Fig.4.19), the gypsum ridges at the north-western end of the area
(1.7), and the escarpments of the fluvial terraces and mesas (2.1).
Fig. 4.19: Residual soil above a siltstone layer. 77
P R O F I L E
PHYSIOGRAPHIC POSITION: Mesa (mapping u n i t 1.1)
LOCATION: P inso ro
DATE: May 24, 1972
LAND USE: i r r i g a t e d c r op s ; sugar b e e t and maize
PARENT MATERIAL: r ed o ld a l luv ium
TOPOGRAPHY: n e a r l y l e v e l
DRAINAGE: w e l l - d r a i n ed
SALINITY: f r e e
DIAGNOSTIC HORIZONS: o c h r i c ep ipedon; cambic ho r i zon and c a l c i c h o r i zon
CLASSIFICATION: C a l c i x e r o l l i c Xerochrept
80
20 cm very dark brown (7.5 YR 4/4) sandy clay; moderate fine crumb structure;- friable (moist); common pores; common very fine roots; clear and wavy boundary
20 - 45 cm very dark brown (7.5 YR 4/4) sandy clay; weak coarse subangular blocky structure; firm (moist); common white lime concretions; some very fine roots; clear and wavy boundary
4 5 - 8 0 cm dark brown (7.5 YR 4/4) and strong brown (7.5 YR 5/6) sandy clay; moderate coarse subangular blocky structure; friable (moist); white lime concretions; gradual smooth boundary
80 - 110 cm dark yellowish brown (10 YR 4/4) clay loam; weak coarse subangular structure; friable (moist); lime concretions
110 - 150 cm brownish yellow (10 YR 6/6) silt loam; structureless; friable (moist); many lime concretions; rootless
P R O F I L E A N A L Y T I C A L D A T A
Depth
cm
0 - 2 0
20 - 45
45 - 80
80 - 110
MO - 150
Sand
2 - 0 . 05
46 .0
49 .5
50 .0
43 .5
12.0
S i z e
T o t a l
S i l t
0 . 05 -0 .002
15.5
13.5
13.0
18.0
63 .5
c l a s s a nd
Clay
<0.002
38 .5
37.0
37 .0
38 .5
24 .5
p a r t i c l e d
Medium coarse
2 - 0 . 25
4 .1
3.9
3.6
6 .9
0 .1
i
s
am
a
J t e r (mm)
n d
F i n e -ve ry f i n e
0 . 25 - 0 . 05
41 .9
45 .6
46 .4
36.6
11.9
Coarse f ragments
>2
% 14.0
10.0
16.0
18.0
-
Carbonate a s CaCOa
<2 mm
% 30 .3
30.2
42 .8
54.4
6 4 . 3
Organic m a t t e r
Z
1.8
1.8
0 .7
0 .4
0 . 3
H20
7.6
7.6
7.6
7 .8
7 .9
pH
6 .9
6 .9
6 .9
7.0
7.1
81
P R O F I L E
PHYSIOGRAPHIC POSITION:
LOCATION:
a l l u v i a l - c o l l u v i a l v a l l e y (mesa /wi th in ) (mapping u n i t 3 .9)
El Bayo
DATE: March 14, 1974
LAND USE: d ry farming ( b a r l e y )
PARENT MATERIAL: mixed l o c a l a l luv ium
TOPOGRAPHY: n e a r l y l e v e l
DRAINAGE: w e l l - d r a i n e d
SALINITY: s l i g h t i n deeper h o r i zon s
DIAGNOSTIC HORIZONS:
CLASSIFICATION:
o c h r i c epipedon
Typic Xerof luvent
82
30 cm brown (10 YR 5/3) sandy clay loam; moderate fine granular structure; friable (moist); abundant fine roots
30 - 65 era yellowish brown (10 YR 5/4) clay loam; weak subangular medium blocky structure; firm (moist) few roots
6 5 - 100 cm brown (10 YR 4/3) silty clay; weak subangular blocky structure; firm (moist); very incipient clay illuviation
100 - 150 cm brown (10 YR 5/3) clay loam; structureless
P R O F I L E 3 A N A L Y T I C A L D A T A
Depth
cm
Sand
2-0.05
S i z e
T o t a l
S i l t
0.05-0.002
c l a s s and
Clay
<0.002
p a r t i c l e d i a m e t e r (mm)
S a n d
Medium Fine-coarse very f ine
2-0.25 0.25-0.05
Coarse fragments
>2
%
Carbonate as CaC03
<2 mm
%
Organic matter
%
0 - 3 0 30 - 65 65 -100 100-150
49.0 43.5 16.5 23.0
27.0 27.0 42.0 41.0
24.0 29.5 41.5 36.0
8 . 2 1.0 0 . 6 0 . 9
40.8 42.5 15.9 22.1
41.5 37.5 35.2 41.5
1.0 0 . 9 1.4 0 . 7
Depth
cm
0 - 3 0 30 - 65 65 -100 100-150
Ex
Ca + Mg
7 .2 7 . 5
13.6 11.1
t r a c t a b l e b a s e s
Na
meq/100 g
0 . 8 1.9 2 . 8 2 . 2
K
0 . 2 0 . 2 0 . 2 0 . 1
Exchangeable so
dium
% 9 . 8
19.8 16.9 16.4
pH
H20
8 . 2 8 . 2 8 . 0 8 . 0
KC1
7 . 4 7 .4 7 . 2 7 . 3
Chlorides
% 0.007 0.016 0.046 0.069
Water e x t r a c t
Ca + Mg Na
meq/1
9.0 0.8 10.0 14.6 18.0 15.6 22.0 20.1
from sa turf i
Sodium a d -sorption r a t i o
0 . 4 6 . 5 5 . 2 6 . 1
t e d p a s t e
E l ec t r i ca l conductivity
mmhos/cm
0 . 6 2 . 3 3 . 7 4 . 2
83
n n r\ c T T P
P R O F I L E
PHYSIOGRAPHIC POSITION: Middle Aragon r iver deposition t e r race (mapping uni t 3.3)
LOCATION: Monasterio de la Oliva
DATE: June 26, 1975
LAND USE: i r r iga ted barley
PARENT MATERIAL: coarse alluvium
TOPOGRAPHY: nearly level
DRAINAGE: extremely good
SALINITY: free
DIAGNOSTIC HORIZONS: ochric epipedon, cambic and ca lc ic horizon
CLASSIFICATION: Calc ixerol l ic Xerochrept
92
2 5 cm yellowish red (5 YR 4/6) sandy clay loam; very common rounded stones over 5 cm; moderate fine crumby structure-; friable (moist); hard (dry); common fine roots; clear smooth boundary
2 5 - 40 cm yellowish red (5 YR 4/6) sandy clay loam; very common rounded stones; moderate very fine subangular blocky structure; firm (moist); very hard (dry); few roots; abrupt and undulate boundary
4 0 - 55 cm continuous layer of semi-cemented rounded stones over 5 cm
P R O F
Depth
cm
0 - 25
25 - 55
Depth
cm
0 - 25
25 - 55
I L E
Sand
2-0.05
52.0
46.5
pH
H20
7.7
7.8
8
S i z e c
T o t a l
S i l t
0.05-0.002
20.5
21.5
Chlor-
KC1
%
7.1 0.006
7.0 0.002
l a s s
Clay
<0.002
27.5
32.0
C
and p a r t i c l e d i a m e t e r (mm
S
Coarse & Medium
2-0.25
12.7
9.4
Water e x t r a c t
a + Mg Na
meq/1
14.0 1.0
8.0 0.8
a n d
Fine Very f
0.25-0
39.3
37.1
ne
05
A N A L
Coarse fragments
>2
%
from s a t u r a t e d p a s t e
Sodium adsorption
r a t i o
0.4
0.1
E l ec t r i ca l conductivi ty
mmhos/cm
1.2
0.6
Y T I C A L
Carbonate as CaC03
<2 mm
%
34.1
30.3
D A T A
Organic
%
1.8
1.4
93
P R O F I L E
PHYSIOGRAPHIC POSITION:
LOCATION:
Low Aragon r i v e r t e r r a c e (mapping u n i t 3 .2)
Rada
DATE: August 29 , 1976
LAND USE:
PARENT MATERIAL: mixed a l luv ium
TOPOGRAPHY: n e a r l y l e v e l
DRAINAGE:
SALINITY:
e x c e s s i v e l y d r a i n e d , groundwater t a b l e +2.50 m
f r e e
DIAGNOSTIC HORIZONS:
CLASSIFICATION:
o ch r i c epipedon
Typic Xerof luvent
94
2 5 cm pale brown (10 YR 6/3) loam with gravel; weak granular structure; friable (moist); abundant roots; clear smooth boundary
25 - 50 cm pale brown (10 YR 6/3) loam; structureless; friable (moist); abundant pores and roots; abrupt and smooth boundary
5 0 - 7 0 cm layer of cobbles and gravel; abrupt and undulate boundary
7 0 - 8 5 cm brown (10 YR 5/3) fine sandy loam; very friable (moist); abrupt and smooth boundary
8 5 - 150 cm layer of cobbles, gravel and coarse sand
1 50 250 cm brown (10 YR A/3) fine sandy loam
P R O F
Dep th
cm
0 - 2 5 25 - 50 70 - 85 150-250
I L E
Sand
2 - 0 . 0 5
4 1 . 2 4 1 . 0 6 1 . 4 5 1 . 3
9
T o t a l
S i l t
0 . 0 5 - 0 . 0 0 2
4 4 . 9 4 3 . 5 2 7 . 7 3 7 . 8
S i z e
C l ay
<0 . 002
13 .9 1 5 . 5 10 .9 10 .9
c l a s s a n d p a r t i c l e d i a m e t e r
C o a r s e Very c o a r s e
2 - 0 . 5
7 . 3 0 . 6 1.1 1.6
S a
Medium
0 . 5 - 0 . 2 5
6 . 7 3 .1
14 .1 1 5 . 5
n d
F i n e
0 . 2 5 -
16 . 3 2 7 . 8 2 6 . 8 1 8 . 5
!mm)
Very f i n e
0 . 1 - 0 . 0 5
10 .9 9 . 5
19 .4 15 .7
A N A L Y T
C o a r s e f r a g m e n t s
>2
X s o i l
50
I C A L
C a r b o n a t e a s CaC03
<2 mm
% 37 . 4 3 8 . 7 4 2 . 0 3 9 . 4
D A T A
O r g a n i c m a t t e r
%
3 . 2 1.4 1.4 0 . 8
Extractable bases Water extract from Dep th
cm
0 - 2 5 25 - 50 70 - 85 150-250
Ca + Mg Na
meq /100 g
7 .1 0 . 2 3 . 6 0 . 2 4 . 0 0 . 3 5 . 8 0 . 2
K
2 . 2 2 . 2 0 . 2 0 . 1
E x c h a n g e a b l e s o d ium
%
2 .1 3 . 3 6 . 7 3 . 3
Ca + Mg
4 . 7 4 . 5 6 . 5 5 . 8
Ca
4 . 3 4 . 3 4 . 1 4 . 1
Na K
meq/ I
1.1 3 .4 2 . 2 4 . 4 4 . 9 0 . 1 1.5
HCOj
5 . 7 2 . 8 I . I 1.3
l. 1. 1 LUli
CI
I
0 . 0 08 0 . 014 0 . 0 55 0 . 010
U 1 . 1
S0»
0 . 3 2 . 2 3 . 2
p
pH
H20
8 . 5 8 . 4 7 . 9 8 . 2
Sodium a d s o r p t i o n
r a t i o
0 . 7 1.5 1.4 0 . 9
E l e c t r . c o n d u c t . mmhos/cm
0 . 8 1.1 2 . 9 0 . 7
95
P R O F I L E 13
PHYSIOGRAPHIC POSITION: a l l u v i a l - c o l l u v i a l v a l l e y (mapping u n i t 3 .8)
LOCATION: Alera p i l o t f i e l d - c o n t r o l a r e a
DATE: August 28 , 1976
LAND USE: s a l i n e v e g e t a t i o n (suaeda e t c . )
PARENT MATERIAL: s a l i n e - a l k a l i l o c a l on decomposed mudstone and s ands tone
TOPOGRAPHY: n e a r l y l e v e l
DRAINAGE: impe r f e c t l y d r a i n ed , ve ry slow runoff , groundwater t a b l e a t 1.40 m
SALINITY: h igh s a l i n i t y and a l k a l i n i t y c o n t e n t , i n c r e a s i n g w i t h dep th
DIAGNOSTIC HORIZONS: o c h r i c ep ipedon, n a t r i c ho r i zon
CLASSIFICATION: Aquic , G lo s s i c Na t r a r g i d
102
0 - 15 cm pale yellow (2.5 Y 7/3) silty clay; light olive brown (2.5 Y 5/3) when moist; moderate medium crumb structure; adherent (wet); slightly firm (moist); slightly hard (dry); many fine pores; many horizontal medium roots; cracks in 35 cm; clear wavy boundary
15 - 35 cm light yellowish brown (2.5 Y 6/3) silty clay, dark gray brown (2.5 Y 4/2) when moist; strong coarse subangular blocky structure; adherent (wet), very firm (moist); many fine and medium horizontal roots; frequent root pores; salt efflorescences; clear and wavy boundary
35 - 70 cm pale yellow (2.5 Y 8/3) silty clay loam; dark gray brown (2.5 Y 4/2) when moist, with very dark gray brown (2.5 Y 5/3) tongues; moderate coarse prismatic structure; clay and silt skins on prism sides and pores; many fine continuous pores; frequent fine and medium roots; salt efflorescences; clear and wavy boundary
70 - 200 cm light yellowish brown (2.5 Y 6/4) when moist silt loam; structureless; medium gypsum crystals at 85-100 cm; very firm (moist); very fine pores decreasing with depth; compact; at 1.4 m saturated with saline water
> 2 00 cm very pale brown (10 YR 7/4) and white (5 Y 8/2) silty clay loam; structureless; poreless and extremely compact; very hard (dry); impervious
P R O F I L E
Uepth
Sand
cm 2-0.05
13
T o t a l
S i l t
0.05-0.002
S i z e
Clay
<0.002
c l a s s and p a r t i c l e d i a m e t e r
Coarse Very coarse
2-0.5
S a n d
Medium Fine
0.5-0.25 0.25-1
I mm >
Very fine
0.1-0.05
A N A L Y
Coarse fragments
>2
Z s o i l
T I C A L
Carbonate as CaC03
<2 mm
%
D A T A
Organic matter
% 0 - 1 5 15 - 35 35 - 70 70 -200
2.2 51.1 1.6 46.3 1.8 68.0 2.1 73.7
46.7 52.1 30.2 24.2
0 . 3 0 . 1 0 . 3 0 . 1
0 . 2 0 . 2 0 . 1 0 . 1
0 . 8 0 . 4 0 . 3 0 . 3
0 . 9 0 . 9 1.1 1.6
- 2 0 . 8 19 .1 3 4 . 1 4 3 . 5
2 . 5 2 . 5 1.4 0 . 6
Depth
cm
0 - 1 5 1 5 - 3 5 35 - 70 70 -200
E x t r a c t a b l e ba
Ca + Mg Na
meq/100 g
12.0 15.7 14.9 9.5
s e s
K
0.8 0.9 0.5 0.1
Exchangeable sodium
% Ca + Mg
5 1 . 2 4 2 . 8 8 1 . 3 9 1 . 8
Ca
3 7 . 3 3 0 . 2 2 5 . 9 2 2 . 0
Na
119 146 210 313
8 4 0 7
W a t e r e x t r a c t f
K HCOj
raeq/1
0 .1 1.9 2 . 3 1.2 0 . 8
CI
0 . 4 6 0 . 4 7 0 .71 1.18
r o m 1 i
SOi,
2 3 . 0 5 5 . 2 9 1 . 1 7 1 . 9
1 p a s t e
pH
H20
8 . 1 8 . 0 7 . 8 7 . 6
Sodium a d s o r p t i o n
r a t i o
23 31 32 46
7 7 9 3
E l e c t r . c o n d u c t . mmhos/cm
14 . 8 1 6 . 3 2 3 . 2 3 4 . 4
103
P R O F I L E 16
PHYSIOGRAPHIC POSITION: a l l u v i a l fan (mapping u n i t 3 .13)
LOCATION: Sab inar
DATE: J u l y 12, 1972
LAND USE: b a r l e y (poor)
PARENT MATERIAL: s a l i n e f i n e a l luv ium
TOPOGRAPHY: smooth 1 p e r c en t s l ope
DRAINAGE: groundwater t a b l e +1.5 m; s low h y d r a u l i c c o n d u c t i v i t y
SALINITY: h i gh
DIAGNOSTIC HORIZONS: o c h r i c epipedon
CLASSIFICATION: S a l o r t h i d i c Xe ro r then t
108
4 5 cm light yellowish brown (10 YR 6/4) when moist, very pale brown (10 YR 7/3) silty clay loam; weak coarse subangular blocky structure; saline efflorescences; smooth and abrupt boundary
4 5 - 8 5 cm very pale brown (10 YR 7/4) silty loam, light yellowish brown (10 YR 6/4) when moist; moderate fine platy structure; hard (dry); whitish efflorescences; abrupt and smooth boundary
8 5 - 150 cm very pale brown (10 YR 7/4) loam, light yellowish brown (10 YR 6/4), when moist; strong very fine platy structure; friable (moist)
P R O F I L E A N A L Y T I C A L D A T A
Depth
cm
Sand
2-0.05
S i z e
T o t a l
S i l t
0.05-0.002
c l a s s and p a r t i c l e d i a m e t e r (mm)
S a n d
Clay Coarse Fine Medium Very f ine
<0.002 2-0.25 0.25-0.05
Coarse fragments
>2
%
Carbonate as CaC03
<2 mm
%
Organic matter
% 0 45 80
- 45 - 80 -150
12.5 28.5 30.5
56.0 51.0 49.0
31.5 20.5 20.5
_ --
12.5 28.5 30.5
38.5 44.0 44.7
0 . 8 0 . 3 0 . 4
Depth Chlorides
%
Water
Ca + Mg
meq
e x t r a c t
Na
/ I
from s a t u r a t e d p a s t e
Sodium E lec t r i ca l
r a t i o mmhos/cm
0 45 80
- 45 - 80 -150
7 8 8
9 2 5
7 7 8
6 8 0
1 0 0
352 358 554
22 80 10
8 0 2
257 141 205
0 0 7
24 22 28
1 3 8
49 21 31
4 0 1
109
5. Soil salinity
5.1 Origin and localization of the salts
The primary origin of the salinity that affects some soil associations
of the studied area is the intrinsic salt content of the Oligocene-Miocene
mudstone and siltstone from which most of the soil parent materials derived
(Chap.4). These Oligocene-Miocene sediments were deposited under brackish
lacustrine conditions (Chap.2).
Erosion of these saline geological formations and their further trans
port and sedimentation under even more saline conditions (salt concentration
increases with the evaporation of water during the transport and sedimenta
tion processes) have formed alluvium and mixed alluvium-colluvium with an
even higher salt content than the original Tertiary sediments.
Although the salinity of the soils has a primary origin, there has been
some redistribution of the salts and a secondary salinization process has
taken place which has changed the original salinity status.
In previous climatic periods when a net percolation of water occurred,
i.e. when rainfall exceeded evapotranspiration, the upper soil horizons were
leached and the salts accumulated at some depth. The leaching rate depended
on the water-transmitting properties of the soils, the more permeable hori
zons losing their salts more readily.
This natural leaching process no longer takes place under present clima
tic conditions where evapotranspiration exceeds effective rainfall for the
greater part of the year, except for a short period during the rainy season.
The introduction of irrigation, however, has changed the water and salt
112
balances in the irrigated soils, and again net percolation of water and
leaching of salts is occurring.
The relative position of the physiographic units and their mutual
interrelation in regard to water flow have played a leading part in the re
distribution of salts (Chap.3). The salts lost from the upper lands have
accumulated in the lower lands. The transport processes are lateral seepage
and to a lesser extent surface runoff.
These processes are still continuing. The higher physiographic units,
however, still contain large amounts of salts. Although this may not always
be evident at the soil surface.
In the lowlands that lack natural drainage or where drainage is impeded,
the groundwater is fed by lateral seepage and the groundwater table remains
near the soil surface. This has produced secondary salinization due to capil
lary rise of the saline groundwater.
The more severely salt-affected soils are therefore either in water
receiving areas with impeded drainage or in those physiographic units where
the soil parent material has intrinsic salinity and where water-transmitting
properties are so low as to impede leaching.
As might be expected, these saline areas are low-lying in relation to
their surroundings. An exception is formed by the recently formed alluvial
soils of the Aragon and Riguel valleys, which, despite their low position,
are non-saline because their soil parent material was free of salinity. They
have satisfactory water-transmitting properties, a low water table, and
apart from the lower end of the Riguel alluvial plain there is no saline
seepage to affect them.
In summary, the source of salts is the intrinsic salinity of some soil
parent materials in combination with a further movement of salts and second
ary salinization in water-receiving areas lacking natural drainage. Under
irrigated conditions the mobilization and redistribution of salts continues
and salinity increases.
113
5.2 Types of salts and their distribution in the soil profile
The saline soils of the area are characterized mainly by the presence
of sodium chloride. From the analysis of groundwater samples (Chap.3) the
composition of the salts of each soil association could be derived.
The soils of the fluvio-colluvial valleys and alluvial plains of the
Castiliscar basin, contain considerable amounts of magnesium sulphate
associated with calcium sulphate; some sodium sulphate is also present.
The soils of the saline-alluvial valleys and fans of the south-western
piedmont major unit contain not only sodium chloride, but also calcium and
magnesium chlorides which predominate over the sulphates.
In the areas associated with the gypsum formations, calcium sulphate
occurs as a major constituent in conjunction with sodium chloride.
In non-saline soils (EC < 4 mmhos/cm) the calcium plus magnesium
content is in excess of sodium. At higher salinity levels the sodium ion
content increases at a greater rate than calcium and magnesium.
As sodium chloride is the most soluble salt, it is leached in periods
of heavy rainfall and low evapotranspiration and during the irrigation
season. Under arid conditions sodium chloride accumulates in the soil sur
face owing to an upward movement of soil moisture containing dissolved
salts.
The main movement of this cycle takes place in the upper horizons of
the soil profile because the saline soils of the area generally have a low
hydraulic conductivity in the unsaturated zone below 50 cm and are almost
impervious below one metre.
Saline soil layers deeper than one metre - and even shallower in soils
with stratified subsoils impervious for vertical flow - are not affected by
114
the present soil moisture cycle. Where deep percolation water flows laterally
through such layers, however, the seepage water is highly saline. Although
such movements may be extremely slow, their long-lasting effect partly
accounts for the high degree of salinity of these layers.
In the non-saline soils calcium and magnesium are present throughout
the soil profile. In the more developed soils carbonates tend to accumulate
in the subsurface horizons. Their common presence combined with a slight
amount of sodium chloride gives a background salinity level for all soil
associations. This is rarely less than 1 mmho/cm, expressed in terms of
electrical conductivity of the saturated paste (EC ).
The sodium adsorption ratio (SAR) increases in conjunction with the
rise of EC, and a similar trend holds for exchangeable sodium percentage
(ESP). However, the ratios between EC and ESP and between SAR and ESP may
vary considerably. Non-saline alkali soils have not been found and pH-values
greater than 8.5 are uncommon.
5.3 Salinity in the soil associations
The soils of the mesas (1.1) are in general non-saline (Fig.5.1).
The parent material from which they derived was salt free and their natural
drainage prevents any secondary salinization. In small depressions, however,
where the coarse alluvium has been removed and a perched water table fed by
lateral seepage overlies the impervious mudstone, salinization can occur,
although in such cases EC-values are only slightly higher than 4 mmhos/cm.
In the soils of the narrow fluvio-colluvial valleys (3.9)
formed by the water courses flowing from the erosion valleys of the mesas, a
shallow water table can be present at places where drainage is impeded. Mo
derate secondary salinization then takes place due to lateral seepage from
the adjacent high-lying mesas whose transmissivity is high.
1 From a large collection of data, a few have been selected as examples of the salinization problem of each mapping unit.
115
* Water extract from \-\ paste
Fig.5.1. Composition of salts in a mesa soil.
Table 5.1 shows the salinity distribution throughout the soil profile.
A slight increase of the chlorides can be distinguished in the surface hori
zon probably due to the effect of evaporation. In addition, the salinity in
creases in the deeper layers above the mudstone with its intrinsic salinity.
TABLE 5.1 Distribution of the salinity in a soil of a fluvio-colluvial valley, slightly saline phase
Soil depth
cm
0 - 5 0
50 - 100
100 - 150
150 - 200
200 - 250
250 - 300
Chlorides
%
0.052
0.053
0.040
0.033
0.052
0.072
EC
mmhos/ctn
6. 1
4.5
3.5
4.3
5.1
7.4
PH
8.2
8.5
8.5
8.4
8.5
8.5
NOTE: Groundwater table at 0.3 m, EC =6. S rmihos/cm
116
The soils of the stony ridges (1.5) are salt-free. Natural
drainage is adequate and salts, even though initially present in the coarse
colluvium, have been leached by rainwater. Even in irrigated soils no
salinity is observed.
The soils of the piedmont slopes (2.3) and colluvial slopes
(2.7) are generally slightly saline,though non-saline or moderately saline
patches could be included in the mapping units. Leaching occurs in these
soils but they also receive salts from higher levels. Thus their actual sal
inity status depends on their relative topographical position: they can be
net losers or net receivers of salts. They are generally somewhat saline at
depth (Fig.5.2) and can therefore be regarded as internal solonchaks.
depth. meq/l.
* Water extract from M paste
Fig.5.2. Composition of salts in a soil of the colluvial slopes.
The soils of the upper and middle alluvial plain of the
Riguel river (3.5) are not salt affected,except at depths of two or more
metres where a slight salinity can be observed. Table 5.2 shows the increase
in the salt content with depth. An explanation of this could be that the
117
• CI" 0SOJ ^HCOi Q c a " | 1 M , " Q No* * water extract from 1 1 paste
Fig.5.4. Composition of salts in a soil of the fluvio-aolluvial valley, saline-alkali phase.
The differences in surface structure development give different possib
ilities for natural leaching. In fact, the extremely saline patches observed
correspond with soils that are more compact than those in adjacent less
saline patches. It is not possible to map with accuracy areas of high or
low salinity because variations can occur over very short distances (see
Chap.6).
Figure 5.4 shows an increase of the most soluble salts with depth,
mainly sodium chloride and some magnesium sulphate. In non-cultivated soils,
however, high concentrations of salts can occur in the surface soil, which
may be covered by a thin layer of salt efflorescence or by a salt crust.
Their presence is due to the fluctuating water table which is a constant
supplier of salts, in combination with high evapotranspiration rates which
occur even in the cold season because of the dryness, intensity, and
frequency of wind.
120
In the subsurface layer a secondary accumulation of calcium sulphate
is found (Fig.5.A). At depths between 80 and 120 cm a horizon with gypsum
crystals ("punta de lanza") is observed (Fig.5.5). It coincides with the
zone in which the water table fluctuates.
The mobility of the soluble and even less soluble salts through the
surface soil and the existence of soil structure or stratification develop
ment in the upper 80 cm are useful indications of the possibilities of re
claiming these soils by leaching and artificial drainage.
Fig. 5.5: Accumulation of gypsum crystals (Profile 14).
Sodification in most soils is no problem because although sodium ions
are present in the clay complex, they are easily exchanged by calcium, which
is present in large quantities. Even in irrigated soils with simple drainage
facilities, no true alkali soils were found but only saline-alkali soils.
This indicates that the leaching of soluble salts is invariably followed
by a corresponding decrease in exchangeable sodium. In fact, the less saline
soils are rarely sodic and pH-values are rarely greater than 8.5.
121
Consequently, with higher leaching amounts supplied by the irrigation
system, and an improvement in the drainage system a decrease in harmful
salts could be expected. The exchangeable sodium on the exchange complex
may well be replaced by calcium ions.
Most of the soils of the alluvial valleys (3.11) and allu
vial fans (3.13) of the south-western piedmont unit are severely affec
ted by salinity. Table 5.3 shows an overall increase of the total salt
content from the upper to the lower valleys.
TABLE 5.3 Electrical conductivity values (EC x 10 ) , chlorides in % e
and pH of three soil sites in the Valarena alluvial valley
u
dep
cm
0 -
3 -
20 -
30 -
50 -
60 -
65 -
80 -
130 -
P P
th
3
20
30
50
60
65
80
130
150
e r v
Cl"
% 0.690
0.325
0.275
0.157
0.230
0.140
0.312
0.242
0.142
a 1 1 e
EC
mmhos/cm
30.7
14.8
13.6
11.3
14.5
11.0
17.2
15.3
10.1
y
pH
H20
7.9
8.2
8.1
8.3
8.2
8.4
8.2
8.1
8.2
M i d
depth
cm
0 -
30 -
40 -
60 -
70 -
90 -
30
40
60
70
90
150
d 1 e
Cl"
% 0.325
0.225
0.325
0.425
0.382
0.410
v a 1 1 e
EC
mmhos/cm
13.7
12.5
18.3
19.7
17.3
21.3
y
pH
H20
8.1
8.3
8.1
8.2
8.3
8.1
L o w e r
depth
cm
0 - 4 0
40 - 75
75 - 85
85 - 115
115 - 150
Cl
% 0.077
0.460
0.165
0.390
0.245
v a 1 1 e
EC
mmhos/cm
-25.2
12.8
24.5
12.8
y
pH
V 8.3
8.3
8.5
8.5
8.5
The salts are rather uniformly distributed in the soil profile (Table
5.3) because the water movement is mainly restricted to the upper 50 cm,
particularly where stratified layers occur.
During dry periods, however, the surface of these soils is covered
with a salt crust (Table 5.3; 0-3 cm depth) which is dissolved again when
rain falls. When the soil surface dries again a very thin crust of platy
structure is formed which breaks out in hexagonal polyhedrons (Fig.5.6).
These soils thus have an uneven surface in which small salt-enriched puffs
(EC up to 68 mmhos/cm) can be found (Fig.5.7).
122
Fig.5.6 a,b: Surface arust of a saline soil.
The salts usually present are sodium chloride and, in appreciable
quantities, sulphates and chlorides of calcium and magnesium. On the soil
surface, patches of darker hygroscopic salt efflorescences can be distin
guished from the light-coloured crust of sodium chloride (Fig.5.8).
123
Fig.5.7: Salty puffs in a soil of a saline alluvial valley.
124
Fig.5.8: Darker patches of hygroscopic salt efflorescences.
Where the water-transmitting properties of the unsaturated zone allow
rain water to pass through the soil, there has been a downward movement of
soluble salts. In such cases the surface soil is slightly saline and inter
nal solonchaks may occur (Table 5.4).
TABLE 5.4 Distribution of the salt content in a soil of the saline alluvial fans
Depth
cm
0-50
50-100
100-150
150-200
200-250
250-300
CI
%
0.122
0.244
0.468
0.536
1.088
1.052
EC
mmhos/cm
7.7
12.8
21.0
22.2
40.9
38.5
PH
8.3
8.1
8.1
8.0
8.0
8.0
NOTE: Water table at 1.25 m, EC = 60 mmhos/em
The possibilities of reclaiming these alluvial soils by leaching with
irrigation water and providing an artificial drainage system are highly va
riable, depending as they do absolutely on the water-transmitting properties
of the soil. Soil with compact and platy or stratified layers would be the
most difficult to reclaim because water passage through the soil profile is
impeded. The data of Table 5.4, however, indicate favourable leaching in
the upper one metre and thus give a more optimistic picture.
5.4 Local crop tolerance to salinity and alkalinity
Several crop tolerance tests were conducted on the saline soils of the
area to compare their results with the generally accepted values for the
tolerance of field crops to salinity and alkalinity.
125
2) Soils slightly affected by salinity (EC = 4 to 8 mmhos/am)
in the vootzone but with increasing salinity at depth.
Comprising this class are the soils of the colluvial (2.7) and piedmont
slopes (2.3).
3) Soils highly affected by salinity (EC > IS mmhos/am)
This class includes the soils of the saline alluvial valleys (3.11)
and fans (3.13) of the south-western major physiographic unit and the soils
of the fluvio-colluvial valleys (3.8), gypsum valleys (3.7), and alluvial
plain (3.10) of the northern drainage basin.
Fig. 5.12: Present salt accumulation in a soil with impeded drainage.
130
The soils classified under 1) can be reclaimed simply by maintaining
the present drainage system, which consists of open ditches and interceptor
drains laid in the contact between the slope and the low-lying valley, and
by applying full irrigation. The percolation losses normal with basin irri
gation are sufficient to leach the salts deposited by secondary salinization
in the rootzone. Where moderate salinity (EC = 8 to 15 mmhos/cm) affects
some soil patches, sugar beet is a suitable reclamation crop because of its
relatively high tolerance level and its high water requirements.
In the soils of the slopes, classified under 2), severe problems
arise when the land is levelled for basin irrigation. Deep saline and less
pervious layers are exposed and the original natural drainage deteriorates.
A second problem is that, after irrigation, saline water seeps from the
slopes to adjacent low-lying areas and increases the original salinity of
the flat alluvial soils.
Irrigation of the slopes should be by sprinkler since this does not
require the land to be levelled. The soil profile is thus not truncated
and the natural drainage remains unchanged. Moreover, better water manage
ment is possible due to better control of the amounts of water applied.
In this way salts can be leached to deeper layers and the seepage of saline
water reduced. The only drainage system required is an interceptor drain
between"the slope and the valley. As the impervious layer in the valley
soils is shallow, an open ditch 1.5 to 2.5 m deep will suffice to catch
the seepage flow.
The alluvial soils classified under 3), which lack natural drainage
and are highly saline, will require a much more rigorous reclamation. Before
they can be used for the full production of irrigated crops, they must be
provided with a drainage system and be subjected to initial leaching.
131
!'he Aiera experimental fieLd represents the poorly drained soils of
the saline fLuvio-coiluviai valleys (.3.8) and the alluvial plains (3.10) of
the northern drainage basin. These soils have an intrinsic salinity because
the mixed aLLUVium-coiLuvium derives from the denudation of the saline
Tertiary muds tones. They have also been subjected to secondary saiinization
owing to the presence of a shallow saline groundwater table. Soil profile
!3 (Ghap.V; is situated in this experimental field.
The V'alarena experimental field represents the saline soils of the
aiiuvial valleys ('3 M ) and tans (3,13.) of the south-western area. Here,
soil :;ai. unity has a primary origin only; the alluvium was derived from ero
sion of the saline mudscone and vas deposited in a stratified fashion in a
brackish fluvial environment. Soil profile !5 (Chap.4) is situated in this
experimental field.
The Alera experimental, field was initially 7.5 ha but one year later it
was increased to 10 ha; in the extension new drainage materials of corrugated
V'VC pipe pre-wrapped with cocos and esparto fibers were used.
The V'alarena experimental fieid was originally 11 ha but was increased
by a further 9 ha to enable the study of a new drainage system designed in
accordance with groundwater flow conditions observed over a period of 2
years on the original drainage system.
6.3 Detailed soil survey
6.3.1 Outline of the survey
In each of the experimental fields representing the two types of saline
soils a detailed soil survey was made. The objectives of this survey were
to test the homogeneity of soil conditions and to determine the soil hydro-
logical properties, the initial salt content, and its variability.
Soil observations were made by means of bore-holes reaching the less
pervious layer (about 2 m deep). The observation points were distributed on
!34
a grid system, the average density of the network being 6 sites'/ha in the.
Alera field and ^ sites/ha in the Valarena field, where soil salinity of the
surface soil was more uniform.
The soi~ survey included descriptions of a representative soil profile
toi each unit and four deep bore holes reaching the underlying Tertiary mud-
stone (from 3 to 5 n; deep). In the Afera field the soil samples taken with
the drilling machine below the groundwater table were completely disturbed,
so three extra deep bore holes were made by hand. The deep scii samples were
analysed to determine the salt content oi deep layers. Piezometer tubes were
installed at different depths.
The hydraulic conductivity of the soils of the. Alera field could be
measured by the auger-hole method, because the groundwater table was only
about ! m deep. At the Valarena field, where the inverse auger hole method
(Porchet) was used, the groundwater table was between 2.5 m and 3 m.
The infiltration rate was measured by the double ring method. The
moisture-holding capacity was estimated from pF-curves derived from soil
core samples.
6 3.2 Hydrologica) soil properties
Only those soil properties affecting land drainage and leaching are
considered in this section. The general characteristics of the soils can be
found in tne descriptions of the mapping units (Chap.4).
Alera experimental field
The Alera experimental field is representative of the. fiuvio-colluvial
valley:-. (Profile 13, the soil is an Aauic Glossic Natragid). The soils are
fine-textured, from silty clay to silty clay loam, though silty loam layers
with a high content of very fine sand produced by the weathering of silt-
stone occiii interbedded between finer lavers,
135
Porosity decreases gradually with depth with a consequent increase in
soil compactness and a decrease in hydraulic conductivity (Table 6.1).
Below an average depth of 1.5 m the soil layers become less pervious,
though they seem to be saturated with water. At a depth variable from 3 to
4 m a mottled Tertiary mudstone is found. The texture of this mudstone is
similar to that of the overlying soil layers, but it is extremely compact
and impervious.
When the detailed survey was made the depth of the groundwater table
varied from 1.0 to 1.2 m. The measured K-values thus refer to the soil
layer between those depths and the bottom of the auger holes (1.5 m ) . Later,
when the drainage system was installed and the initial leaching process had
started, the hydraulic conductivity values of the surface soil layers were
measured during periods of high water table.
TABLE 6.1 Hydraulic conductivity values obtained by the auger-hole method (Alera experimental field)
0
0
1
Depth (m)
0-0.50
50-0.75
75-1.00
00-1.50
Texture
silty clay
silty clay loam
silty clay loam
silty loam (very fine sand)
Structure
moderate blocky
weak blocky
structureless; gypsum crystals
structureless
Consistency (moist)
moderately firm
firm
firm
friable
K (m/day)
0.6"
0.3-0.4
0.2
0.1-0.2
K-values measured by the inversed auger hole method in the surface soil
were approximately zero.
The hydraulic conductivity of undisturbed soil cores determined in the
laboratory (Table 6.2) revealed that the soil has an anisotropic permeabil
ity, the vertical percolation rate being much greater than the horizontal
hydraulic conductivity. The downward decrease of hydraulic conductivity was
reaffirmed, the layer below 1.5 m being almost impervious, even for vertical
flow. Where the surface soil had been subsoiled, however, a relative increase
was evident.
136
TABLE 6.2 Hydraulic conductivity values from laboratory determinations in undisturbed soil cores (Alera experimental field)
Depth (m)
0-0.50
0-0.30
0.50-0.85
0.85-1.00
0.85-1.00
> 1.50
Horizontal-K (m/day)
0.004
0.040
0.010
0.040
-0.001
Vertical-K (m/day)
0.97
0.69
0.73
0.28
2.50
0.02
R e m a r k s
-subsoiled surface layer
--
gypsum c rys ta l layer
compact subsoil
The absolute K-values determined for horizontal hydraulic conductivity
were much lower than those obtained by the auger hole method.
The infiltration rate was initially estimated in a ploughed soil which
was later left uncultivated. The measured values varied from 15 to 20 mm/day.
For the initial drainage design the Tertiary mudstone was considered
the impervious barrier, at an average depth of 3.5 m. With a drain depth
of 1.5 m and assuming a hydraulic conductivity of 0.1 m/day for the layer
between 1.5 and 3.5 m, the mean transmissivity value below drain level 2
would be 0.2 m /day.
5 -
3-
2-
1 -
C 1
10 1
2 0
— 2 5 - 50 cm
\1
\ \ 1 1 1 30 4 0 50 Vol %
Fig.6.1: pF-ourves of the soils of the Alera experimental field.
137
From pF-curves (Fig.6.1),the drainable pore space (y) and the available
moisture content between field capacity and wilting point could be derived.
The moisture content under saturated conditions (pF = -°°) could be
assumed approximately equal to the moisture content 1 cm above the ground
water table under equilibrium conditions (pF = 0), because in soils where
macropores are absent the pF-curve is almost vertical for pF-values close
to zero. If under equilibrium conditions the moisture content 100 cm above
the groundwater table (pF = 2) corresponds with field capacity,the drainable
pore space can be estimated from the difference between the moisture content
by volume at pF = 2 and pF = 0.
According to this assumption the estimated drainable pore space of the
upper 1 m averages 3.7 per cent. The estimated average available moisture
content (between pF=2.0 and pF=4.2) in the upper 0.5 m is 7 per cent.
4 ,00—
0,50—
0,75—
1,75—
3,00
\ \ \ ' • / /
2.5 Y 4 / 2 . S i l t y c lay . Modera te blocky s t ruc tu re . Mode ra te l y f i r m
2.5 Y 5/4. S i l t y clay l o a m . Poor s t r u c t u r e . F i r m .
10 YR 5 / 6 S i l t y clay. S t r u c t u r e l e s s . Very f i r m
2.0 Y 5 /6 . S i l ty l oam ( ve ry f i n e s a n d ) . S t r u c t u r e l e s s . F r iab le
10 Y R 5 / 6 . S i l ty c lay l o a m . S t ruc tu re less , Extremely compac t and f i r m
Muds tone S i l t y c l a y - s i l t y c lay loam. Hard and d ry
Fig. 6.2: Schematic soil profiles. Alera experimental field.
138
The variation of soil conditions can be observed in Fig.6.2 where
a schematic description of two characteristic soil profiles is shown, and
in Table 6.3 where the corresponding particle size distribution data
are listed.
TABLE 6 . 3 T e x t u r a l a n a l y s i s of s o i l p r o f i l e s of F i g . 6 . 2
D e e p b o r e h o l e - b a s i n 1
Depth
m
0-0.50
0.50-0.75
0.75-1.00
1 .00-1. 50
1.50-2.50
2.50-4.00
4.00-5.00
gravel
>2 mm
-------
2-1
--
1. 121
----
1-0.5
-0.311
2.241
----
s
0
a
.5-0.25
0.20
0.361
2. 131
0.52
2.05
0.66
1.08
n
0.25-0.
0.46
0.3I1
1.76
3.78
4.26
3.50
3.90
d
25 0.125-0.05
1 .24
0.67
2.66
15.30
4.77
5.89
9.87
silt
0.05-0.002
55.91
50.85
52.10
56.06
54.80
46. 11
46.50
clay
<0.002 mm
42. 15
47.47
37.95
24.27
34.00
43.82
38.63
D e e p b o r e h o l e - b a s i n 2
Depth g r a v e l s a n d s i l t c l a y
m >2 mm 2-1 1-0.5 0 . 5 - 0 . 2 5 0 . 2 5 - 0 . 1 2 5 0 . 1 2 5 - 0 . 0 5 0 . 0 5 - 0 . 0 0 2 <0.002
0 - 0 . 5 0 - - - 0 . 50 0 .35
0 . 5 0 - 0 . 7 5 -
0 . 7 5 - 1 . 7 5 - - - 0 . 7 0 ' 0 . 3 5 1
1 .75 -2 .25 -
2 . 2 5 - 3 . 0 0 -
>3 .00 - - -
1.41
0.73
0.351
0.99
5.18
1.56
51.51
69.73
52.92
66.92
52.25
50.91
46.20
29.53
45.66
32.07
42.55
47.52
gypsum crystals
139
Valarena experimental field
The Valarena experimental field is situated in a saline alluvial valley
(3.11). The soils are fine-textured in the upper 2.5 m, the texture varying
from silty clay to silty clay loam. Below 2.5 m on average, the very fine
sand content increases and a silty loam texture is found.
These fine sediments overlie a gravel and coarse sand layer which has
a variable thickness of 1 to 1.5 m. In the experimental unit the gravel
layer is continuous and its upper limit is at a depth of 3 to 3.5 m. When
the detailed soil survey was made (June 1974) the gravel layer was satura
ted with very saline water (Chap.4, Profile 15; the soil is a Salic Stratic
Xerofluvent).
This aquifer lies above a red and grey mudstone whose depth varies
from 4 to at least 5 m. The texture of the mudstone is silty clay and, in
spite of underlying a saturated zone, is apparently dry. For this reason it
can be considered the impervious barrier.
The upper soil layers show a marked very fine stratification. The
hydraulic conductivity is therefore anisotropic, the horizontal hydraulic
conductivity being higher than the percolation rate.
The hydraulic conductivity was measured by the inverse auger hole method
because the water table was deep. An average K-value of 0.2 m/day was ob
tained from 12 measurements made in a soil layer situated between 1.0 and
1.5 m (future drain level). In the surface soil where the stratification had
been disturbed by land levelling and further by subsoiling and where there
was some halophytes root profileration, a mean K-value of 0.5 m/day was ob
tained.
Later, during the initial leaching process, attempts were made to de
termine the hydraulic conductivity by the auger hole method. No results could
be obtained, however, as no true water table occurred in the stratified
layers.
Assuming a constant K-value of 0.2 m/day in the layers above the gravel 2
layer, the transmissivity below drain level would be 0.4 m /day. No direct
hydraulic conductivity measurements were made in the gravel layer, but its
140
transmissivity is assumed to be higher than 1 m /day.
The infiltration rate as determined by the double ring method averaged
10 mm/day in the unploughed surface soil.
In Fig.6.3 a schematic soil profile is shown, the textural analysis
of which is given in Table 6.4.
^50-
/, 10 YR 5 /4 . Silty cloy loam. Structureless
10 YR 5/4. Silty clay. Very fine stratification
10 YR 5/4. Silty loam (very fine sand). Structureless
. Gravel and coarse sand (saturated with water)
Red mudstone. Extremely hard ana dry
Fig.6.3: Schematic soil profile. Valarena experimental field.
TABLE 6.4 Particle size distribution of soil profile of Fig.6.3
Depth
m
0-0.50
0.50-1.00
1.00-1.50
1.50-2.00
2.00-2.75
2.75-3.25
3.25-3.50
coarse f ragm.
>2 ram
--
- -----
2-1
-------
1-0.5
0.06
0.06
0.11
0.12
0.30
0.10
3.72
s
0.5-0
--
0.05
0.06
0.10
0.10
1.18
a
25
n
0.25-0.
0. 12
0.06
0.16
0.06
0.10
2.37
3.88
d
25 0.125-0.05
1.02
0.32
0.62
0.48
0.45
10.84
11.59
silt
0.05-0.002
70.30
58.10
62.98
54.48
52.73
55.58
55.86
clay
<0.002 mm
28.47
41.43
36.15
44.78
46.29
30.97
23.73
141
6.3.3 Initial salinity
Alera experimental field
Soil salinity increases with depth, being highest in the less per
vious layers (> 1.5 m), where the electrical conductivity (EC ) varied from
20 to 35 mmhos/cm (Table 6.5).
In the surface soil the variation was greater, with slightly saline
soil patches (EC -4 mmhos/cm) close to very saline soils (EC >30 mmhos/cm)
Though soil samples were taken on a grid system and the observation
density was 4 sites/ha, it was not possible to produce a map with different
intervals of soil salinity because variation was so great over short
distances.
TABLE 6.5
Depth
m
0-0.50
0.50-0.75
0.75-1.00
1.00-1.50
1.50-2.50
2.50-4.00
4.00-5.00
0-0.50
0.50-0.75
0.75-1.75
1 .75-2.25
2.25-3.00
>3.00
Salinity
col
%
25.8
35.5
36.2
41.5
38.3
36.5
35.4
23.1
44.2
30.3
39.1
35.2
30.1
of the soil
Saturation moisture content
7.
62.5
68.0
61.0
40.5
51.0
58.4
52.0
64.0
49.0
75.0
54.0
64.0
70.0
layers of Fig
Organic matter
%
3.2
1.4
0.9
0.8
0.9
0.9
1.2
3.2
0.7
0.7
0.9
1.0
1.1
6.2 (February
H20
7.8
7.7
7.6
7.8
7.9
7.8
7.7
8.0
7.7
8.0
8.0
7.9
7.8
PH
KC1
7.2
7.2
7.3
7.3
7.5
7.3
7.0
7. 1
7.2
7.2
7.8
7.4
7.4
1976)
Cl"
% 0.06
0. 13
0.12
0. 12
0.27
0.54
0.55
0.017
0.028
0. 170
0.340
0.380
0.470
Ca++& Mg+
meq/1
17.7
48.4
43.3
29.0
55.6
82.3
96.3
13.4
13.7
45.9
47.8
64.0
54.8
Na +
meq/1
50.0
100.4
100.4
106.8
161 .5
243.0
267.0
19.9
30.7
110.0
187.0
175.0
193.0
SAR
16.8
20.4
21.6
28.0
30.6
37.9
38.5
7.7
1 1 .7
23.0
38.3
30.9
36.9
ESP
19
22
24
29
31
35
35
10
14
25
35
31
35
EC e
mmhos/cm
6.I1
12.41
12.4
13.0
19.8
28.9
32.7
2.91
4.01
12.4
22.8
22.3
23.6
Layers leached with irrigation water
142
The salinity levels were reflected in the sodium adsorptium ratio and
under equilibrium conditions the soils must be assumed to be saline-alkali.
Gypsum crystals were evident in the fluctuation zone of the water table.
At the start of leaching the salinity of the groundwater was high.
Though the salt content varied slightly, from the evidence of water sample
analysis (Table 6.6) it could be observed that highly soluble salts, NaCl,
MgSO, and Na.SO, were the most commonly occurring. The groundwater appeared
to be saturated with CaSO,.2 H„0 (40 meq/1). Where the total salt content
was very high (Drain No.20), some CaCl„ and MgCl„ occurred (Fig.6.4).
The gypsum content was not determined because of the particular diffi
culty of such an analysis in these saline soils. However, a high gypsum
content was assumed because of the occurrence of gypsum crystals in the
fluctuation zone of the groundwater table (0.85 to 1 m) and because the
groundwater was saturated with CaSO,.2 HLO. Thus it seemed likely that the
soluble salts could be leached without gypsum applications.
TABLE 6.6 Groundwater initial salinity. Alera experimental unit
Drain
No.
1
3
7
8
15
18
20
EC
mmhos/cm
60
48
46
50
43
46
85
Salt
g/1
37.5
29.6
29.6
32. 1
26.6
29.4
59.7
Cl"
470
365
350
435
315
425
950
SO*
145.8
116.6
131.6
104.1
116.6
73.3
81.2
HC03
3.8
3.7
3.6
3.7
4.6
3.0
3.4
Ca+ +
m e q / 1
36.0
31.6
46.8
46.8
43.4
44.4
68.0
Mg
56.0
40.6
50.2
57.0
44.6
54.8
139.0
K+
0. 1
0.1
0.0
0.1
0.0
0.0
0.1
Na+
535
425
400
430
360
400
822
pH
7.3
7.8
7.6
7.6
7.7
7.4
7.3
SAR
78.9
70.7
57.4
59.7
54.3
56.8
80.8
143
meq/1.
-1.000
- 5 0 0
- 300
- 2 0 0
100
1 ' sea water
2'. drain n?1
x — 3. drain n? 20
4! drain n?15
Fig.6.4: Sahoeller diagram for groundwater. Alera experimental field (December 1974).
144
Valarena experimental field
Soil salinity affecting these alluvial soils was more uniformly distri
buted than in the soils of Alera. There was a slight increase in the salt
content of the surface layers when evapotranspiration was high,but the sali
nity of the deeper stratified silty clay layers was more homogeneous (Table
6.7).The surface crust of uncultivated surface soil (<5 cm) had a high salt
concentration (EC - 70 mmhos/cm) and a pseudostructure with saline puffs.
TABLE 6.7 Salinity of soil layers of Fig.6.3 (June 1976). (Water extract from saturated paste.)
m
0-0.50
0.50-1.00
1 .00-1.50
1 .50-2.00
2.00-2.75
2.75-3.25
3.25-3.50
C03
%
43.2
39.4
41.8
37. 1
35.0
42.5
44.0
Saturation moisture content
%
57.5
65.0
62.5
65.0
72.5
47.5
44.5
Organic matter
%
3.3
1.6
1.5
1.5
1.3
1.0
1.3
H20
7.5
7.6
7.6
7.8
7.8
7.8
7.8
pH
KC1
7.3
7.4
7.5
7.6
7.6
7.5
7.5
CI
%
1.09
1.08
1.07
0.95
0.82
0.66
0.62
Ca++& Mg+
meq/1
243.3
190.5
183.8
164.3
144.3
191.8
163.0
Na+
meq/1
307.0
325.0
355.0
325.0
275.0
307.0
307.0
SAR
27.8
33.3
37.0
35.8
32.4
31.3
34.0
ESP
29
32
35
34
31
31
33
EC e
mmhos/cm
49.0
43.7
46.3
40.4
33.5
39.3
40.0
Though salinity did not vary over such short distances as in the Alera
unit, the filled in part of the levelled terrace close to the collector
drain was less saline than the excavated part. This indicated that where the
very fine stratification is altered by levelling operations and further by
subsoiling, water passes more easily, and leaching is enhanced.
The sodium adsorption ratio increased with salinity, making the saline
soil alkali as well.
The groundwater which saturated the gravel layer above the impervious
mudstone was very saline, but there was a great difference in the total salt
content of two water samples taken at two different sites.
The composition of the salts was similar to that of the groundwater of
the Alera field,NaCl and MgSO, being the most common (Table 6.8 and Fig.6.5).
145
:.0CO !
5 0 0 !
»0 -j
»oo -I
, 0 0 i /
50-
/
CO-/
1 ' <;eo wafer
2 c - 7 0
0,5- h- 0,5
0,3-1 -Q3
0,1 -
Co ' Mq S O . HCO'
Scioeller -j-'i-ajram for grvunchnatei'. Jal.weha -jx;:eri'
146
As some dispersion was observed in the surface soil crust, a gypsum
application seemed a reasonable method of preventing alkalinization during
leaching.
TABLF 6.8 Salt composition of the groundwater. Valarena experimental unit
EC Salt C) SCK HCO? Ca Mg K Na pH
mmhos/cni g/i m t' q / J
C-70 66 39.1 585 80., 3.8 62.4 89 0.1 520 7.3 59.8
C-59 28. ', 18.7 262 45.2 7.2 25.2 20 0.0 27 1 7.4 57.0
147
7. Experimental field design
7.1 Design of the drainage system
The theoretical calculation of the drainage system was based upon the
most commonly used drainage equations for steady and unsteady-state condi
tions. To these equations, simplified and assumed drainage criteria and the
hydrological soil properties measured by conventional methods were applied.
7.1.1 Alera experiments 1 field
Hydrological soil conditions
A drainable soil profile with the average hydrological soil properties
determined in Section 6.3.2 was considered (Fig.7.1).
silty clay K=Q6 m/day
A = 3.7% p silty clay K=0.2-0.4 m/day
silty loom— K=0.1-0.2 m/day silty clay loam
silty clay - K=0.I m/doy silty cloy loom
^ ^ V \ \ \ ^ ^ ^ ^ ^ N S \ V \ \ \ \ ^ S \ \ \ \ ^ X \ ^ mudstone- impervious layer
Fig. 7.1: Drainable soil profile. Aleva experimental field.
148
Although the hydraulic conductivity decreases gradually with depth,
the boundary between the layers with K = 0.1 m/day and K = 0.2 m/day was
considered to be at a depth of 1.5 m. In calculating the drain spacing this
boundary was chosen as the average drain depth.
Drainage criteria
Different drainage criteria were formulated for steady and unsteady-
state conditions, giving separate consideration to the irrigation season
and the winter period when no irrigation takes place.
For the design of the drainage system the initial leaching period with
out crops was not taken into account, but only the control of the saline
groundwater table once the rootzone (the upper 1 m soil) is free of salts.
Drainage criteria for the irrigation season
The irrigation water losses percolating through the unsaturated zone
cause a recharge of the groundwater. The rise of the groundwater table
depends on the amount of percolation water and the drainable pore space (y).
The drainage system had to prevent a gradual rise of the water table
during the irrigation season. The field application losses could be estimated
from irrigation experience in the area. In the peak irrigation period appli
cations of 80 mm every 12 days are common. Since the land is levelled, sur
face losses are negligible. If the field application efficiency is e =0.7,
the percolation losses are:
_ „ v 80 mm 24 mm „ ,, s = 0.3 x ,„ , = -r=—3 = 2 mm/day
12 days 12 days
These losses could be expressed in steady-state conditions as a con
stant recharge. The drain discharge rate should then be 2 mm/day, with
the water table remaining at a depth of 1 m (h = 0.5 m and s = 2 mm/day).
149
Instead of steady-state equations, however,it is more logical to apply
unsteady-state equations, since percolation losses cause a sudden rise
of the water table followed by a gradual fall. With u = 3.7% (see 6,3.2),
the initial rise of the water table should be:
. 24 mm n ,. Ah = 7T „ = 0.65 m
U . U j /
Thus a drawdown of 0.65 m is required in a period of 12 days. To obtain
a water table that is comparable with the steady-state water table at a
depth of i m (h = 0.5 m), the following values were assumed as criteria for
unsteady-state flow (h = 0.825, h = 0.175, t = 12 days):
Fig. 7. 2: Drainage criteria for the irrigation season. A'lera experimental field.
From the depth-duration-frequency relation derived with the Gumbei
probability distribution (Chap.2) a rainfall of 68.5 mm can be expected
over 3 consecutive days during the irrigation season for a return period
of 5 years. This amount of rainfal'j is less than an irrigation application.
For this reason the formulated criteria remain valid, if the possibility of
heavy rainfall after an irrigation application is neglected.
Although the salinity of tne irrigation water is low (EG =0.3 mmhos/cm)
the leaching requirements during the irrigation season could be calculated
150
using cue sale equilibrium equation (van der Mo!en and van Hoorn, 1973):
e
jhere
R = leaching requirements (mm)
S = amount of evapotranspiration (mm)
P = effective amount of precipitation (mm)
EC.. = mean electrical conductivity of the irrigation water (mmhos/cm)
t = leaching efficiency coefficient with respect to percolation
EC - maximum value of the electrical conductivity of the e
S O L ! saturation extract (.mmhos/cm;
!'he precipitation deficit was calculated from the evapotranspiration
data. The latter were calculated with the Thornthwaite and Blaney-Criddle
methods and in accordance with the Penman method, using meteorological data
from 2aragoza (Chap.2). The total amount of orecipitation was regarded as
effective rainfall because no surface runoff is expected on flat levelled
1 ana.
if we assume values of f - 0.5 and EC = 2 mmhos/cm, the leaching
requirements calculated by means of Sq.(') are ''22 mm, the corresponding
-heoretieal net irrigation requirements Deing 873 mm. Annual irrigation
applications of 1000 mm are common in the area, so if the field application
efficiency is 0.', aormal percolation losses are approximately 300 mm and
are thus in excess of the leaching requirements.
the drainage criteria adopted for steady-state conditions are similar
to the drainage requirements applied m some irrigation projects in arid and
jemi-arid zones, for instar.ee those reported from the Medjerda Valley in
Tunesia i,s = 2 mm/day, water table depth = ; m) and the Habra Valley in
Algeria (s = 2 mm/day, depth - 0.8 m ) , as well as those recommended for the
oardenas area by Hr..v'ardie and Dr Ede (personal communication).
151
Drainage criteria for the winter season
During winter, when there is no irrigation, the groundwater is re
charged by rainfall and possible seepage from the adjacent slopes.
Maximum rainfalls derived from the depth-duration-frequency curves
(Chap.2) were 76 mm over 3 consecutive days and 85 mm over 6 consecutive
days for a 5 year return period.
Assuming that the unsaturated zone has a moisture content near field
capacity, a percolation of 40 mm would be sufficient to cause the water
table to rise to the surface, even though initially the water table was
deeper than 1 m.
The drainage system must therefore be capable of obtaining a water
table drawdown of 30 cm in 3 days while at the same time maintaining a
constant discharge rate equal to the recharge intensity. This recharge can
be approximately equal to the basic infiltration rate (about 15 mm/day).
In extreme conditions waterlogging occurs for 3 days, with a drainage dis
charge of 45 mm.
In the hydropedological survey no measurable seepage was noted, which
is not surprising in view of the low transmissivity of the soils of the
slopes. The steady-state summer drainage criteria (s = 2 mm/day, h = 0.5 m)
could therefore meet any possible seepage requirement.
35 ^ ^ ^ ^ ^ ^ ^ ^ ^ ^
Fig. 7. 3: Drainage criteria for the winter season. Alera experimental field.
152
The winter drainage criteria could thus be formulated as follows:
s = 15 mm/day, h = 1.5 m for steady-state conditions and h = 1.5 m, h
1.2 m, t = 3 days for unsteady-state conditions (Fig.7.3).
The drainage criteria are summarized in Table 7.1.
TABLE 7.1 Drainage criteria. Alera experimental field
_ Water flow Season ... h s h h t
conditions o t (m) (mm/day) (m) (in) (days)
Irrigation steady state 0.5 2 - - -
unsteady state - - 0.825 0.175 12
Winter steady state 1.5 15
unsteady state - - 1.50 1.2 3
Drain spacing
The Hooghoudt theory was used to calculate the drain spacing for
steady-state conditions. Though the hydraulic conductivity decreases
gradually with depth, the drain level was considered to be at the boundary
between two layers with different hydraulic conductivities.
For the flow region above the drain level an average hydraulic
conductivity was adopted.
Applying the formulated drainage criteria (Table 7.1) and the measured
hydrological factors (Chap.6) to the Hooghoudt Equation (2) the drain
spacings of Table 7.2 were obtained.
8 K, dh + 4 K h2
L2 = — i * (2) s
153
where
L = drain spacing (m)
s = discharge rate (m/day)
K = hydraulic conductivity of the layer above drain level (m/day)
tC = hydraulic conductivity of the layer below drain level (m/day)
d = equivalent depth of aquifer below drain level (m)
h = hydraulic head above the drain level midway between drains (m)
TABLE !.2 Drain spacing, obtained with Eq.(2). Alera experimental field
Drainage cr i ter ia Hydrological factors Drain spac ing 1
h s K K, D
O.) (m/day) (m/day) (m/day) (m)
0 . 5 0 .002 0 .2 0 .1 2
1.5 0 .015 0 .4 0 .1 2
L o
(m)
22 .4
20 .0
c o
(m)
5 .5
5 .5
L
(m)
16.9
14.5
iei'd : ~.J?C): L = L - c , c = D In — , u = TTr , r = 0 .04 m
It was not easy to select an equation to calculate the drain spacing
Ecr unsteady-state conditions. The Glover-Dumm equation was not applicable
because the region of flow above drain level is more important than the 2
region below (d = i.1 m, Kd = 0.11 m /day). Besides, neither the thickness
of the aquifer nor the hydraulic conductivity in the drainable soil profile
is constant.
The Boussinesq equation seemed a better fit if the water flow below
drain level was not taken into account and if the hydraulic conductivity
for the region of fLow above drain level was considered constant. This
equation then reads:
4.46 K h h t L2 5 _ L _ (3)
u(hQ - ht)
where
h •= hydraulic head above drain level after an instantaneous recharge (m)
h = hydraulic head above drain level midway between drains at any time t (m)
t = time (days)
y = drainable pore space
TABLE 7.3 Drain spacing obtained with Eq.3. Alera experimental field
Dra inage criteria
h o
(m)
0.825
1.50
ht (m)
0.175
1.2
Hydrological
t
(days)
12
3
K
(m/day)
0.4
0.4
factors
V
0.037
0.037
Dr ain spacing
L
(m)
11.3
29.5
From the calculated drain spacings of Tables 7.2 and 7.3 it could be
concluded that a drain spacing of between 15 and 20 m was suitable. The spac
ing finally selected for testing in the Alera field was 20 m for an average
drain depth of 1.5 m (Fig.7.4). Part of the undrained control area is shown
in Fig.7.5.
Drainage and filter materials
Smooth clay pipes and corrugated PVC pipes, 8 cm in diameter, were used
as field drains.
The hydraulic design of the smooth clay pipes was made with the Wesse-
ling equation (Eq.4), and that of the corrugated pipes with the Manning
equation (Eq.5)
QT = s B L = 8 9 d 2 - 7 1 2 r ° - 5 7 2 (4) Li
. ,Q ,2.667 .-0.5 ,,, Q = s B L = 38 d l '.5j
where
3 Q = discharge rate (m /day)
s = discharge rate per unit surface area (m/day)
155
Fig. 7. 4. Partial view of the Aleva experimental field.
Fig. 7. S. Undrained control area. Alera experimental field.
156
2 B x L = dramable area (m )
B = drain length (m)
d = drain internal diameter (m)
i = hydraulic gradient = drain slope
If the maximum specific discharge is 15 mm/day and the drainable area 2
5 200 m (L = 20 m, B = 260 m ) , an internal drain diameter of 8 cm and a
drain slope of 0.1% is sufficient for both the clay and PVC pipes. For the
first installation, no 8 cm PVC pipes were available and 5 cm pipes were
used. For the comparative study of the entrance resistance of drainage and
filter materials, the following combinations were selected (Fig.7.6):
Drain Group 1: clay pipes (d = 80 mm, length 40 cm)
Drain Group 2: clay pipes as group 1 with a gravel cover (gravel tf = 2 - 20 mm, 25 1/m, Fig.7.7)
Drain Group 3: corrugated PVC pipes (d = 50 mm) with a gravel cover as group 2
Drain Group A: corrugated PVC pipes (d = 50 mm)
Drain Group 5: corrugated PVC pipes (d = 50 mm) with barley straw cover
Drain Group 6: pre-enveloped corrugated PVC pipes (d = 50 mm) with esparto fibre (Fig.7.8)
Drain Group 7: pre-enveloped corrugated PVC pipes (d = 50 mm) with cocos fibre
Drain Group 8: as drain group 2.
Collector drain and main drain
The simple drainage system consists of (Fig.7.6):
a) Field laterals installed at a 20 m spacing.
b) Two open collector drains which collect water from the field laterals (D-IX-0 for drains 1 to 15 and D-IX-3-2 for drains 16 to 26, Fig.7.9)
c) Main drain which is an existing open ditch (C-XVIII,Fig.7.10)
d) The existing open collector D-IX-3 which is used as an interceptor drain
e) A new open ditch was designed to collect the surface drainage of basins 1 to 5 (Fig.7.11).
157
E E E E F
• (M lO ^ i n (O N o
! ! ! i !
S N I V M O 1 9 H U J 1
: « JCE
£
O
3 O
«
158
_ , . j a w -" •"• •M**""*
Fig. 7. 7: Installation of cZ-ay drains.
Fig'. 7. 8: Installation of plastic pipes pre-enveloped with esparto fibre
159
Fig.7.9: Field drain outlets to collea- Fig. 7.10: Main drain of the tor dravn. Alera experimental field. Alera field.
The main drain has a trapezoidal cross-section with a bank slope of 1:1
and an average depth of 2 m. The bank is severely eroded and is partly co
vered by halophytes (Suaeda and Atriplex).The salinity of the drainage water
is such that it prevents the growth of aquatic weeds so that water is free
flowing.
The collector drains debouch into the main drain through concrete
culverts built below the road C-IX-1 (Fig.7.12).
160
' - ! : & & & * • • * * • • • •
***e*ft
Fig. 7.11: Open ditch to collect surface drainage. Alera field.
Fig.7.12: Concrete culvert below road.
The hydraulic design of the open co l lec tor d i tches was calculated
with the Manning equation (Eq.6):
^ A R ^ S ' / 2 (6)
where
Q = KM = A =
R =
S =
design discharge (m /s)
Manning coefficient (m 2
wetted area (m )
hydraulic radius (m)
hydraulic gradient
1/3 Is)
161
The design discharge could be calculated because the maximum spec.ifi'_
discharge (s = lb mm/day) and the maximum drainable area (7.8 ha) were known.
The discharge rate thus obtained was i 3.5 1/s (q ~ 1.74 1/s/ha).
If the lateral outlets are at s c.epth of 1.6 m and the safety margin
to the water level in the ditch is 0.1' m, the water: in the ditch has to be
at a depth of at least: 1.8 m.
Assuming a Manning coefficient of 40 in' /s and a hydraulic gradient
of 0.05 per cent (to correspond with the slope of the land), the following
section factor was obtained for the design discharge:
A R 2 / 3 = 0.0,5 m 8 / 3
A trapezoidal cross-section with a bank slope of 1:1, a bottom width
b = 0.4 m, and a water depth in the ditch y = 0.2 m, has a section factor 8/1
of 0.03 m ", which is twice the required value.
The flow velocity would be v = Q/A = 0.1 m/s, which is less than the
maximum permissible velocity in a silty clay soil (v = 0.7 m/s).
An open-ditch collector-drain thus has the following characteristics: 2
a trapezoidal cross-section (w.4 m*~), a bank slope of !:1, bottom width
0.4 m, depth 2 m, and slope 0.05%
7.1.2 Valarena experimental field
Hydrologxcal soxl conditions
The. average hydraulic conductivity of the stratified silty clay layers
was K = 0,2 m/day ''inverse auger hole method). The hydrauHc conductivity
of the deep gravel, layer was not measured, but was assumed to be at leas!"
i m/day.
162
Some laboratory measurements were made on undisturbed soil cores
ind showed Lower EC-values than those obtained in the field.
Che drainable pore space was estimated by a ruie-of-thumb developed
by van Beers (1962)' When the hydraulic conductivity has been determined
and the K-vaiue is expressed in cm/day, the root of the K-value is appro
ximately equal to the drainable pore space (in X) . Lt was thus found that
u - 4.5%.
An average drajnabie soil profile with the average soil hydrological
properties is as shown in Fig./.i3.
si l tv c lay loam K^ OD m/day
s t r a t i f i e d :-'.ii!y cloy loam
K = 0.2 m/day
yU.- 4 . 5 %
si l ty l o a m - f i n e sand
grave l and sand K * l m/day
-nudstone - imperv ious layer
Fig.'/. IS: Drainable soil profile. Valarena experimental field.
The boundary between the saturated gravel layer and the underlying
reddish dry mudstone was regarded as the impervious barrier.
The design drain depth was chosen at 1.5m because the nearer it was
to the pervious gravel layer the better for drain installation.
163
Drainage criteria
The steady-state drainage criteria formulated for the Alera field were
also applied here. Unsteady-state drainage criteria were not formulated
because of the difficulty of applying unsteady-state equations in a strati
fied soil.
Drainage criteria for the irrigation season
Although the intensity of heavy rainfall increases from north to south,
the maximum rainfall to be expected over 3 consecutive days is 75 mm for
a 5 year return period (Chap.2). This amount of water is comparable to an
irrigation application, so the drainage criteria (s = 2 mm/day, h = 0.5 m)
are valid.
As at the Alera field, the leaching requirements are less than the
percolation losses.
Drainage criteria for the winter season
In winter the water table can rise to the soil surface because 85 mm
and 93 mm of rainfall are expected respectively over 3 and 6 consecutive
days in a 5 year return period (from depth-duration-frequency curves, Chap.
2). The discharge rate under steady-state conditions must therefore be equal
to the basic infiltration rate (s = 10 mm/day) for a hydraulic head h = 1.5 m.
The drainage criteria for steady-state conditions are summarized in
Table 7.4.
TABLE 7.4 Drainage criteria. Valarena experimental field
Season
irrigation
winter
Water flow conditions
steady state
steady state
h
(m)
0.5
1.5
s
(mm/day)
2
10
164
Drain spacing
The drain spacing for steady-state flow was calculated with the Ernst
formula because of the variability of the hydraulic conductivity due to
stratification and also because the design drain level does not coincide
with the boundary between the layers of different hydraulic conductivity.
Applying the drainage criteria and the hydrological factors to the
Ernst equation (Eq.7), the drain spacings of Table 7.5 were obtained. They
refer to summer and winter conditions, respectively.
D L2 h = S i; + S 8 (KjD, + K2D2) + S L Wrad ( 7 )
where
h = total hydraulic head midway between drains (m)
s = drain discharge rate per unit surface area (m/day)
L = drain spacing (m)
K = hydraulic conductivity of the upper less pervious layer (m/day)
K„ = hydraulic conductivity of the more pervious layer (m/day)
D = thickness of layer over which vertical flow is considered (m)
D = average thickness below the water table of the upper layer with hydraulic conductivity K (m)
D„ = thickness of the lower layer with hydraulic conductivity K„ (m)
W = radial resistance (days/m) rad
The radial resistance could be calculated with Eq.(8) because the
design drain level is above the more pervious layer and because K„/K. < 20:
K. W = K. W1 + | In (8) 1 rad 1 TT 4u
where
a = geometry factor (m)
u = wet perimeter (m)
165
By means of the Ernst nomograph for determining the radial resistance,
which is derived by numerical calculation, a W .-value of 5.5 days/m
was obtained.
TABLE 7.5 Drain spacing obtained with Eq.(7). Valarena experimental field
Season
summer
winter
Drainage
h
(m)
0.5
1.5
criteria
s
(m/day)
0.002
0.010
Hydrological factors
K! K2 (m/day)
0.2 1.0
0.2 1.0
(m)
1.25
1.25
D 2 (m)
1.75
1.75
D V
(m)
0.5
1.5
Dra in spacing
L
(m)
33.0
.21.0
As the winter conditions are more restrictive, a drain spacing L = 20 m
with drain level at 1.5m was adopted.
Drainage and filter materials
Except for the pre-enveloped PVC pipes with esparto and cocos fibre,
the same drainage and filter materials as used at Alera field were installed
at Valarena (Fig.7.14).
Pipes 8 cm in diameter and a drain slope of 0.1% were selected because,
although the design discharge rate is less than that of Alera, the drainable
area is similar.
Collector drain and main drain
The open ditch D-XX-8 collects the water flowing from the field drains
and debouches into the Valarena stream, which is the main drain of the
irrigation sector.
The depth of the collector drain (> 2 m) is sufficient to allow a
safety margin (> 30 m) between the lateral outlets and the water level in
the ditch.
166
Si
i r t i r s i ; 5 5 S s i a i i
1 »
s i 8 s 6 2 S
SUflTMO TTO31T1
.h
'a
(Si
<3 IS <» !H
a
o +i s o «
en
167
7.2 Design of the irrigation system
Since the Bardenas area was planned to be irrigated by gravity flow,
the same irrigation method was used in the design of the two experimental
fields, though sprinkler irrigation would have offered some advantages due
to better control of irrigation applications.
The method chosen for the initial leaching process was intermittent
basin irrigation. Once the desalinization process had been completed, barley
and wheat could be irrigated by basin flooding, maize and sugar beet by
furrow irrigation, and lucerne by borderstrip irrigation.
7.2.1 Alera experimental field
The irrigation system consists of an existing rectangular concrete
irrigation canal, A-IX-4-2. Its total height is 0.40 m and its bed width
0.50 m. The discharge capacity of the canal is 50 1/s. It has four outlets
for basins 1 to 4 and ends in a tail escape into the main drain (Fig.7.6).
A new irrigation canal was designed to irrigate basins 5 to 8 (Fig.7.15).
This canal, A-IX-4-2-1, was calculated with the Manning equation (Eq.9):
2/3 1/2 Q = K^ A R ' S " (9)
where
3 Q = discharge (m /s)
1/3 K^ = Manning coefficient (m /s)
2 A = cross sectional area (m )
R = hydraulic mean depth (= area/wetted perimeter, m)
S = bed slope 2/3 8/3
AR = section factor (m )
168
Fig.7.IS: Irrigation canal. Alera experimental field.
1/3 If for a rectangular concrete canal a coefficient K^ = 66.7 m /s is
2/3 8/3
assumed and the design bed slope is 0.1%,a section factor AR = 0.024 m
will be sufficient for a discharge capacity Q = 50 1/s. A rectangular canal
with a bed width of 0.5 m and a design water height of 0.25 m has a section 8/3
factor of 0.03 m , which is greater than required. With a freeboard of 0.15 m, a canal bank equal to 0.4 m is adequate.
This irrigation canal debouches into the collector drain,D-IX-3-2,through
a concrete tail escape to a stilling basin at the level of the ditch bed.
The distributary canal outlets have a metal weir with a concrete bed
protection downstream of the bed.
For the initial leaching, the land was graded with an almost zero slope.
The largest irrigation basins cover an area of 1.5 ha (length = 300 m,
width = 50 m. Two transversal checks divide the basins into three smaller
ones 0.5 ha) for better water distribution.
169
In addition, an observation well was installed midway between two
drains (10m) to determine the fluctuation of the watertable and to check
the piezometer readings of the tubes placed at that site.
Three extra piezometer lines were installed with tubes at different
depths to determine the hydraulic heads of any aquifers affecting the
drainage system (Fig.7.18).
' * ~* " h^l i« • * *"•"**•— --*
Fig.7.17: Piezometer tube on a elay drain.
Fig.7.18: Installation of a piezometer line with tubes placed at different depths.
172
At the Alera field the inlets of these piezometers were at the follow
ing depths: 1.5 m (drain level), 2.5 m (less pervious layer), 3.5 m (on the
impervious layer)), and 5 m (below the impervious layer).
At the Valarena field the inlet depths were: 1.5 m (drain level),
2.5 m (fine stratified layer), 3.5 m (gravel layer), and 5 m (below the
impervious layer).
The piezometers are iron tubes with an internal diameter of 3 cm; their
lower 10 cm is perforated (Fig.7.19).
disturbed soil
concrete seal
disturbed soi I
gravel t i l ter
.1! •:-:-i I xl tx:
"1 p :1k
• gravel f i l te r
Fig. 7.19: Sketch of piezometer and observation well.
Drain discharge
Discharges were measured by hand readings with a 5 1 bucket in periods
of high discharge and with a 1 1 bucket when discharges were lower.
173
Irrigation flow measurement
The irrigation flow was measured with a metal Parshall flume placed
in the irrigation canal (Fig.7.20). The maximum free-flow capacity of the
flume was 80 1/s; it had a throat section W = 15 cm and a total height
E = 45 cm.
Fig. 7. 20: Parshall flume for irrigation flow measia-ement.
174
Soil sample analysis
Once the initial salinity status of the surface soil was known,7 sites
with different salt content were selected at Alera and 11 at Valarena to
determine the variation of the soil salinity during the leaching process.
At each site soil samples were taken in the following manner. In a 2 .
square 9 m in extent 9 auger holes were made. For the upper 25 cm, a
soil sample was taken by mixing the soil of all the 9 holes. For the 25-50
cm and 50-100 cm layers, soil samples were taken by mixing the soil of 5
holes; 5 holes were considered sufficient because the variation of salinity
was less in the deeper layers than in the surface soil.
Of the selected soil samples the salt content of the saturated paste
was periodically determined in terms of electric conductivity (EC ) , per
centage of chlorides, Na , Ca , and Mg contents, and pH.
Afterwards the field salinity was monitored with soil resistance
measurements (EC-probe and four-electrode equipment set up in the Wenner
arrangement).
Gypsum treatments
To find out whether gypsum improved the surface soil structure and con
sequently increased the infiltration rate, gypsum tests were conducted on 2
small plots ( 1 2 m ) where amounts of 5 and 10 t/ha were applied.
Meteorological station
Daily data on rainfall, temperature, and evaporation were collected at
the Oliva Abbey meteorological station. As this station is 10 km from the
Alera experimental field a rainfall gauge was installed at Alera to obtain
the local rainfall figures.
At Valarena daily data on rainfall and temperature were collected at
the station located there.
175
Unfortunately it was not possible to obtain an anemometer, hygrometer,
and solarimeter in time to allow the calculation of evapotranspiration by
the Penman method.
7.5 Field observation programme
Daily readings were made of piezometric levels, drain discharges, and
meteorological data. As drain systems with a length of 250 m have two
piezometer lines, each line was read every second day.
To observe the rate of groundwater rise during irrigation applications
and the reaction of the drainage system to an instantaneous recharge, the
piezometer lines and drain discharges were measured two or three times a
day in such periods.
In addition, the infiltration rate during irrigation applications was
measured with simple rings placed at several points in each basin and in
the gypsum test areas.
During the initial leaching process soil samples were taken after
each intermittent leaching phase, just before the next irrigation was due
and when the soil was sufficiently dry to allow good mixing. Once the
leaching process had been completed the soil sampling was done seasonally.
176
8. Derivation of agro-hydrological factors
8.1 Groundwater flow conditions
The conditions under which the groundwater flows towards the drains
were determined by daily measurements of the water table and of the drain
discharge. Since these conditions differed in the two experimental fields,
they were analysed separately.
8.1.1 Alera experimental field
The groundwater flow was studied in the clay tile drains with gravel
cover (Drain Groups 2 and 8). These drains produced the clearest data be
cause they had the lowest entrance resistance and consequently gave the best
drainage performance.
Hydrographs of drain discharge (s in mm/day) and hydraulic head midway
between drains (h in m) were drawn for several separate periods. From this
fairly extensive material a selection has been made to illustrate groundwater
flow conditions. Figure 8.1 refers to Drain Group 2 in January-February
1975 during which a leaching application and rainfall recharged the ground
water. Figure 8.2 refers to the same drain group in July-August 1975. The
analysis was repeated for another part of the field (Drain Group 8) in
January-March 1976, when the groundwater was recharged by heavy rainfall
after a leaching application (Fig. 8.3).
177
s(mm/day)
I.I
lo
os
es-
Q7-
06-
0 5 -
04-
0 3
0.2
0.1
f ~ \ s(t)
-13
42
lr 130mm.
20
Jan. R R R _R
2 m m 20mm. 12.5mm 9.5mm.
20
Febr. 1975 R lr
5mm. 130mm.
Fig.8.1: hit) and sit) hydrographs after a leaching application and subsequent rainfall in winter (Drain Group 2, clay/gravel, January-February 1975. Alera experimental field).
s(mm/qoy)
30 1 Sep.
R 14mm
5
R 4 mm
Fig.8.2: hit) and sit) hydrographs after a leaching application in summer (Drain Group 2, clay/gravel, July-August 1975. Alera experimental field).
178
h(
10-
09-
0 *
07
0.6-
05-
0.4-
03-
0 2
0.1
O
m)
J i
soil surface
Cs
\ ^ \ \ \ \ \ \ \ \ \ \
\
\y
\
l \ \ 1 \ \
\ \ . . N \
sit)
- - ^
s(mm/doy)
10
9
8
7
6
5
4
3
• 2
1
0 15 20
Febr 1976
10
March R
48.5mm
Fig. 8. 3: hit) and sit) hyapographs after a leaching application in winter, followed by rainfall (Drain Group 8, clay/gravel, January-March 1976. Alera experimental field).
The hydrographs demonstrate that unsteady groundwater flow prevailed.
In periods without recharge, the water table remained slightly above
drain level (h = 0.1 to 0.2 m). When recharge occurred after leaching or
heavy rainfall, it caused an instantaneous rise of the water table. After
leaching applications of 100 to 150 m», the water table rose to within 10
cm of the soil surface (h = 1.1 to 1.3 m) and remained there for 1 or 2 days
as long as the recharge equalled or exceeded the drain discharge. After
the recharge ceased, the water table gradually fell, and at the end of tail
recession conditions resembled steady flow. In reality, however, the flow
was unsteady since both drain discharge and hydraulic head continued to
decrease with time.
A similar pattern appears in the discharge hydrographs. Immediately
after an instantaneous rise of the water table, the drain discharge changed
from about 0.1 or 0.2 mm/day at conditions approximate to steady-state flow
to about 25 mm/day. As long as infiltration continued, this heavy discharge
179
would last (usually for 3 or 4 days), after which it decreased gradually
until it reached its minimum rate at the end of tail recession.
After the initial leaching period had ended, the fluctuation of the
water table and the discharge were studied under normal irrigation in summer.
Figure 8.4 shows the hydraulic head and discharge hydrographs for August-
September 1976. The amounts of irrigation water varied from 85 to 100 mm
and were applied at an average interval of 14 days. Percolation losses
from 30 to 35 mm are normal with surface irrigation, which means a field
application efficiency e of 0.7 (see Chap.9). These water losses caused
the water table to rise to about 50 cm below the soil surface, at which
time the discharge was about 15 mm/day, decreasing to 0.5 mm/day at the
end of tail recession.
Mm)
12
l.t-
10-
0 9
oe
0.7
0 6
0 5
0 *
QJ
0 2
0 *
(V
tl
/
; /
\ .
soil surface
t l t 1 I I 1 1 I I J |
***""*--... /
11 \ i > \ i > \
V\ 1 drain level
T\ I \
\ \ V \ \ \ \ \
\ V 1 J
s(mm/day)
\ ! ]r\ ( \
1 \ \ \
\ \
H3
H2
-II
-10
9
8
-7
•6
•5
-4
- 3
2
: l
-n 10 12 14
t lr=96mm. R»20mm.
Aug 1976
IS 20 22 24 26 28 30 t t t
10 12 14 16 18 20 22 24 t t I t
"84mm R=10mm.R=5mm. If«90mnv
26 28 30 2
Sep.
R*5mm. Oct.
Fig.8.4: hit) and sit) hydrographs during the irrigation of sugar beet (Drain Group 2, clay/gravel, August-September 1976. Alera experimental field).
180
Table 8.1 summarizes water t able depths, hydraulic heads, and d i s
charges during the i n i t i a l desa l in iza t ion process and subsequent i r r i g a t i on
seasons. This table i s based on a greater amount of data than shown in Figs.
8.1 to 8.4 and covers the whole observation period.
TABLE 8.1 Water t able depth, hydraulic head, and discharge. Alera experimental f ie ld
Initial
leaching
Irrigation
of sugar beet
Water table
initial
m
0.1 0.3 0.5 0.7 1.0 1.1
0.5 0.7 1.0
depth
final
m
0.3 0.5 0.7 1.0 1.1 1.2
0.7 1.0 1.2
h o
m
1.3 1.1 0.9 0.7 0.4 0.3
0.9 0.7 0.4
Unsteady state
h t
m
1.1 0.9 0.7 0.4 0.3 0.2
0.7 0.4 0.2
t
days
3 2-4 1-2 3-9
6 6
2 2 8
s
mm/day
25 >
8 6 3
0.5 <0.2
15 4 2
72
4 3
0.5 0.2
-4 2
0.8
Steady
h
m
_ ---
0.3 0.2
---
state
s
mm/day
_ ---
0.4 0.2
---
1 initial final
8.1.2 Valarena experimental field
At Valarena the actual flow pattern differed from the assumptions made
in the theoretical design of the drainage system (Chap.7).In reality,the wa
ter flows toward the drains horizontally over a shallow impervious layer
(Fig.8.5).
0.5
1.5 %
subsoiled layer
finely stratified layer impervious to vertical flow.
2.5-confined aquifer of sand and gravel.
'/////////////////////////////////////////////////////, impervious mudstone
Fig. 8.5: Groundwater flow conditions at Valarena experimental field.
181
The hydraulic conductivity of the stratified silty clay layers is
anisotropic. Measured by the inversed auger hole method (Chap.6), their
hydraulic conductivity for vertical flow is much lower than that for hori
zontal flow. This was confirmed in laboratory measurements with undisturbed
soil cores. Water movement through the finely stratified layers can thus
be neglected and the abundant soil pores observed in the soil survey are
obviously discontinuous.
As the impervious stratified barrier is at a depth of 50 cm, the water
flows directly into the drain trench through the upper soil layer where the
stratification has been broken up by subsoiling and levelling operations.
The pervious soil profile is therefore restricted to the upper 50 cm. This
restriction causes part of the irrigation water to flow horizontally over
the soil surface toward the open collector drains.
Obviously, there is no deep percolation of water. In fact, the piezo
meters rarely showed any recharge of the groundwater. Most of them re
mained dry during the entire leaching process,. Others indicated low water
levels (from 1.2 to 1.4 m below soil surface) which presumably owe their
existence to some leakage of surface water through the concrete seal. Minor
fluctuations of the water level were observed in a few piezometers.
The actual hydrological conditions are thus those of a shallow ephe
meral perched water table which is never in contact with the deeper gravel
and sand aquifer which, in its turn, is confined by the overlying stratified
layers.
The drain discharge reflected similar conditions, with water flowing
through the drains only during irrigations. Afterwards the discharge de
creased rapidly and within 4 or 5 days the drains were dry. Another indica
tion that the irrigation water flowed directly into the drains was the low
salt concentration of the drainage water.
182
8.2 Drainage equations
8.2.1 Steady flow
Although, in reality, steady flow did not occur at the Alera experimen
tal field, the fluctuation of the water table and the variation of the drain
discharge were so small at the end of tail recession (Table 8.1 and Figs.
8.1 to 8.3) that in these short periods steady flow can be assumed.
The hydraulic conductivity decreases with depth (Chap.6) but a constant
K-value can be assumed at small heads (h = 0.2 to 0.3 m).
Under these conditions the Hooghoudt theory (1940) can be used. The
discharge is then given by the following equation:
8 1 d 4 K h + £ h2 (1)
where
s = discharge per unit area (m/day)
L = drain spacing (m)
K = hydraulic conductivity above drain level (m/day)
}L = hydraulic conductivity below drain level (m/day)
d = equivalent thickness of aquifer below drain level (m)
h = hydraulic head midway between two drains (m)
From the discharge and hydraulic head hydrographs of the Alera field
a discharge-head relation was obtained (Fig. 8.6). This curve is parabolic,
which shows that the discharge-head relation is not linear but quadratic 2
and means that the term h of Eq.(l) is more important than the term h.
Consequently, the greater part of the water flows above drain level, which
183
indicates that the drains were placed above or near the impervious layer
(4 K >> a
s mm/doy s mm/day
22-
0 0.2 0.4 0.6 0.8 0 01 0.2 0.3 0.4 0.5 0.6 07 0.8 0.9 1.0 1.1 1.2 him)
Fig.8.6: s-h relation (Drain Group 8 elay/gravel, February-March 1976 and Drain Group 2, alay/gravel, August-September 1975. Alera experimental field).
8.2.2 Unsteady flow
Alera experimental field
The theory of Kraijenhoff van de Leur (1958), being based on a gradual
rise of the water table due to a steady recharge until steady-state condi
tions are reached, could not be used because the water table rose instanta
neously.
The theory of Glover-Dumm (1954) seemed more appropriate because it
describes the fall of the water table after an instantaneous rise due to
recharge from heavy rainfall or percolation losses of irrigation water in
an aquifer of almost constant thickness (D).
The Glover-Dumm equation for the hydraulic head midway between two
184
drains in relation to drain level reads:
n-1 4 h °° 2 -n2 t/j
h (L/2, t) = — - 2 - £ (-1) I e (2) TT i , 3 , 5 , . . n
where j is the reservoir coefficient (days)
y L 2
(3)
where
2 KD = transmissivity of the aquifer (m /day)
M = drainable pore space
t = time (days)
h = hydraulic head midway between two drains as a function of t (m)
h = hydraulic head of the initial horizontal water 0 table (m)
During tail recession, the second and following terms of Eq.(2) can be
neglected as their values are comparatively small. Equation (2) then reduces
to:
4 h -t/j -t/j h = — e = 1.27 h e (4)
77 O
In a later publication, Dumm (1960) assumed that the initial water
table is shaped like a fourth-degree parabola. If so, Eq.(4) changes to:
192 ,_ , 2 "t/j -t/j
h = - ^ h (TT - 8) e = 1 .17 h e (5) TT5 °
The general equation for the discharge per unit area is:
8 KD h °° -n2 t/j 2. 2
2 1 » 3 j 5 !
s = - 2 e (6)
185
which can be simplified if terms of higher order are neglected:
8 KD h -t/j o
s = e (7)
During tail recession the hydraulic head and discharge for times
t and t„ are
- t l / j hi = 1.17 h e
8 KD h -ti/h o
si - e L2
-t2/j h2 = 1.17 h e
o
8 KD h -t2/j S2 = — • e
L2
-(t2 - t!)/j h2 = hi e (8a)
"(t2 - ti)/j s2 = Si e (8b)
From Eq. (8) the reservoir coefficient can be expressed as:
log hi - log h2 log si - log s2
4- = 2.3 — = 2.3 (9) J t2 - ti t2 - ti
If the discharge and hydraulic head hydrographs are drawn on semi-loga
rithmic paper, two straight parallel lines must be obtained, making an
angle a with the horizontal axis with
log hi - log h2 log si - log s2
tan a = — = t2 - ti t2 - ti
and the reservoir coefficient
• = 1 J 2.3 tan a
186
t. days.
Fig. 8. 7: hit) and sit) hydrographs showing the inadequacy of Glover-Dumm theory for the Alera groundwater flow conditions.
Figure 8.7 shows a plot of the hydraulic head and discharge of Fig.
8.3 on semi-logarithmic paper. The lines obtained during tail recession
are not parallel. This means that the flow above drain level is much greater
than that below and that the thickness of the drainable profile (D) is
far from constant. It also means that the conditions for which the Glover-
Dumm equation were derived are not fulfilled.
187
191
The Boussinesq equation could then be applied to solve the drainage problem.
Drains 2 - July 1975
Drains 8- Feb-Mar 1976
' Drains 2-Jan-Feb. 1975
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 Mm)
Fig. 8.11: Calaulation of the K-values from the s/h-h relation (Boussinesq equation).
By the method of linear regression, using discharge rates and hydraulic
heads observed in the clay tile drains with gravel cover (Drain Groups 2
and 8), the straight lines s/h = f(h) were calculated (Fig.8. 11). From these
straight lines, K-values were obtained (Table 8.3).
In general the values agreed reasonably well and enabled a mean hydrau
lic conductivity of 0.6 m/day to be derived.
Table 8.3 shows that the calculated K-value (0.76 m/day) is higher than
the average value when the water table is deeper than 0.80 m. The explana
tion for this may be that if h is small, the flow below drain level could
be relatively important. If so, the calculated hydraulic conductivity above
drain level will be higher than the actual value.
194
TABLE 8.3 K-values derived from the s/h-h relation, according to Eq.(15)
Period
Jan.-Febr. '75
July '75
August '76
Jan.-Febr. '77
February '76
July - Aug. '76
Jan.-Febr. '77
June - July '77
No. drain
of group
clay/gravel)
2
2
2
2
8
8
8
8
Drawdown of water table
m
0.50 - 1
0.90 - 1
0.50 - 1
0.80 - 1
0.30 - 1
0.10 - 1
0.60 - 1
0.50 - 1
.10
.20
.20
.20
.10
. 10
.10
.00
Drain depth
m
1.4
1.4
1.4
1.4
1.2
1.2
1.2
1.2
Correlation coefficient
s/h-h
0.96
0.96
0.97
0.97
0.96
0.91
0.97
0.94
tan a
3.78
6.40
7.59
6.70
4.05
8.67
3.81
4.80
X
X
X
X
X
X
X
X
io'3
io'3
io'3
io"3
io'3
io'3
io'3
io'3
K
m/day
0.44
0.74
0.88
0.77
0.47
1.00
0.44
0.55
The hydraulic conductivity of the surface layer (0-50 cm) is higher
than the average value because a higher value is calculated when a shallow
water table occurs. The hydraulic conductivity of the surface layer can vary
from 1 to 1.5 m/day. This reflects the better development of structure in
the surface layers.
Second method
Equation (10) expresses the hydraulic head midway between two drains
(h ) as a function of time (t). With this equation theoretical h -curves can
be calculated for several K-values. By comparing the theoretical curves with
the tail recession curve obtained from the piezometer readings, the actual
hydraulic conductivity value can be found.
Several hydraulic head hydrographs were drawn for periods in which
the initial hydraulic heads midway between drains (h ) differed. The actual
curve was compared with theoretical curves derived for several hydraulic
conductivity values. Figure 8.12 shows the fall of the water table in Drain
Group 2 during January-February 1975; Fig.8.13 shows that in Drain Group
8 during February-March 1976.
195
The average K-value determined during three consecutive winters
(0.6 m/day) did not agree with that obtained for the summer periods
(1.1 m/day). The first value agrees well with the average value (0.6 m/day)
obtained from Table 8.3.
The higher K-values in summer may be explained by capillary rise in
the intervals between water applications. This causes the water table to
fall more rapidly and as a consequence the calculated hydraulic conductivity
appears to be higher than it actually is (Figs.8.15 and 8.16).
—+ actual curve
K =0.6 m/day
K = 1.0 m/day
K = l.2 m/day
11 12 13 14 15 Aug 1976
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 (tdays)
d r a i n l e v e l
Fig. 8.15: Calculation of the K-values with the hit) function (Boussinesq equation) in a period of high evapotranspiration (Drain Group 8, clay/gravel, August 1976).
The use of this (second) method for calculating the hydraulic conducti
vity is limited to periods of low evapotranspiration. Moreover, the drainable
pore space must be fairly constant, because otherwise discontinuities appear
in the hydraulic head hydrograph when the falling water table passes through
a layer of different porosity (Fig.8.17).
198
h(m)
1 actuol curve
K = Q5 m/day
K = 06 m/day
K =0.7 m/day
K =0.8 m/day
- - - • • " K = 0.9 m/day
, — , — , « = i j m/day
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1 July 1975
l e v e l Aug.
2 3 4 5 6 7 8 9 10 (t days)
Fig.8.16: Calculation of the K-values with the hit) function (Boussinesq equation) in a period of high evapotranspiration (Drain Group 2, clay/gravel, July-August 1975).
~~^~~.
1 actual curve K = 0.5 m/day
K = 0.6 m/day
23 24 25 26 27 28 29 I 2 3 4" 5 6 7 8 9 10 H 12 Y5 14 \5 16 17 18 19 20 ((days) Febr. 1975 March.
d r a i n l e v e l
Fig. 8.17: Influence of the \i-value on the calculation of the K-values with the hit) function (Boussinesq equation).
199
Third method
The discharge hydrograph was compared with theoretical discharge curves
calculated from Eq.(11) for various hydraulic conductivity values.
From the h -hydrographs the corresponding theoretical s -hydrographs
were derived. For the same K-value of 0.6 m/day obtained from the hydraulic
head hydrograph (Fig.8.13) and for a depth between 0.9 m and 1.2 m (Fig.8.18),
the actual discharge curve of Drain Group 8 (clay/gravel) fits well with the
theoretical one. Above 0.9 m the actual discharge is higher than the calcu
lated value. This is because the shape of the water table differs from that
assumed by Boussinesq until after the upper part of the saturated soil beside
the drain trench is drained (Fig.8.14). Below a depth of 1.2 m the discharge
curve is discontinuous at the end of tail recession, probably due to a dif
ference between the drainable pore space of the silty loam layer and the
value used in the calculations (Fig.8.19).
s [mmydoy)
actual curve
K = 0.6 m/day
Fig.8.18: Calculation of the K-values from the sit) hydrograph (Boussinesq equation).
200
s (mm/day)
1—- actual curve
K =0.5 m/day
K = 0.6 m/day
20 21 22 23 24 25 26 27 28 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17(t.days) Febr. 1975 March.
Fig. 8.19: Influence of the \i-value on the calculation of the K-values from the sit) hydrographs (Boussinesq equation).
For Drain Group 2 (clay/gravel) the actual discharge curve is parallel
to, but higher than, the theoretical one calculated for the K-value of 0.33
m/day obtained from the h -function (Fig.8.12).The difference of 0.15 mm/day
(Fig.8.20) is probably due to seepage.
s(mm/doy)
— i actual curve K =0.3 m/day
K =0.4 m/day
20 21 22 23 24 25 26 27 28 29 30 31 1 2 Jan.1975 Febr.
3 4 5 6 ( ( d a y s )
Fig.8.20: sit) hydrograph showing the existence of a local, low amount of seepage (Drain Group 2, clay/gravel, January-February 1975).
201
8.4.2 Steady flow: Hooghoudt equation
As explained in Section 1.1, steady flow does not, in reality, occur.
At the end of tail recession, however, because of a small amount of seepage,
the flow of some drains (Group 2) approximates steady flow.
In such periods, i.e. when the fluctuation of the groundwater table is
small and the discharge is almost constant, the Hooghoudt equation can be
used to derive the hydraulic conductivity.
As the thickness of the aquifer below drain level can be neglected, the
Hooghoudt equation (1) can be simplified to
s = (16)
This equation is similar to the Boussinesq expression for the
discharge (Eq. 11). Solving Eq.(16) for K yields:
„ s L2
4 h2
To determine the K-values during the d e s a l i n i z a t i o n process , several
per iods with almost s t e ady - s t a t e condi t ions were s e l e c t ed . The r e s u l t s are
g iven in Table 8 .5 and can be compared with those determined with Eq . ( 11 ) .
TABLE 8.5 K-values derived according to Eq.(16)
mm/day m/day
Pe
3-4
7-8
20-22
27-29
22-23
riod
Febr.
July
Jan.
Febr.
Aug.
'78
'75
'76
'76
'76
2
2
2
2
2
No. of drain group
clay/gravel
clay/gravel
clay/gravel
clay/gravel
clay/gravel
Depth of the layer
m
1.10-1.40
1.15-1.40
1.25-1.40
1.25-1.40
1.25-1.40
0.33
0.26
0. 17
0. 15
0.15
0.37
0.27
0.32
0.25
0.19
0.35
0.40
1.141
1.191
0.841
Some contribution of flow below drain level
202
The analysis with the Hooghoudt equation confirms the results obtained
with the equations for unsteady flow.
If the drains are not located exactly on the impervious layer but some
what above it, the influence of the flow below drain level will show up in
apparently higher K-values for the layer above drain depth. The higher the
h-value used in the calculation, the lower the influence of flow below drain
level and consequently the smaller the error.
8.4.3 Hydraulic conductivity of the less pervious layer
At the end of tail recession the piezometers placed at a depth of
3.5m showed a higher water level than those placed at drain depth. This
difference was observed in the three piezometer lines installed at the
Alera field.
There is no clear explanation for this difference because below drain
level the soil is uniformly compact silty clay down to the impervious mud-
stone and no pervious layers were detected. However, as a low discharge was
observed at the end of tail recession, the hydraulic gradient must have
caused an upward flow.
Various periods could therefore be selected in which the vertical
hydraulic conductivity of the less pervious layer below drain level could be
determined. Darcy's law can be applied to vertical flow:
v = K ' ^ = ^ (.7) z D c
where
v z
flow rate = discharge per unit area (m/day)
hydraulic conductivity for vertical flow (m/day)
203
Ah = hydraulic head difference between the 3.5 m deep layer and drain level (m)
D' = thickness of the layer between 1.5 m and 3.5 m (m)
c = D'/K' hydraulic resistance of the less pervious layer (days)
Table 8.6 shows the results obtained. The average K'-value of the
less pervious layer is about 0.002 m/day, which is at least 10 times smaller
than that of the overlying layer above drain level. The layer below 1.5 m
can therefore be considered impervious for flow towards drains, thereby
confirming the conclusions of the study of groundwater flow.
TABLE 8.6 K-values of the less pervious layer
Period
9-11 Dec.
23-24 Feb.
4-5 Mar.
24-27 Mar.
'75
'76
'76
'76
h3.5'
m
2.68
2.61
2.59
2.57
h1.51
m
2.38
2.40
2.38
2.30
Ah
m
0.30
0.21
0.21
0.27
D'
m
2
2
2
2
s
mm/day
0.40
0.45
0.20
0.17
c
days
750
467
1 050
1 588
K'
m/day
0.003
0.004
0.002
0.001
Reference level = depth of the impervious mudstone (3.S m)
To summarize, a value of 0.6 m/day can be used for the hydraulic con
ductivity of the soil profile between a depth of 0.5 m and drain level.
Below drain level the soil is considered impervious. The permeability of
the upper layer (0-50 cm) is about 1.5 m/day. Hence, if the water table
is shallow after heavy rainfall or irrigation, a value of 1 m/day can be
used for the hydraulic conductivity of the soil profile.
204
8.4.4 Comparison of the results obtained
During the experimental period the hydraulic conductivity for different
depths of the groundwater table was measured by the auger hole method. The
hydraulic conductivity above the water table was estimated by the inversed
auger hole method. In addition, undisturbed soil cores were taken in hori
zontal and vertical directions to measure the hydraulic conductivity in the
laboratory.
The object of these direct measurements of hydraulic conductivity was
to compare field and laboratory methods in which only a small body of soil
is measured, with large-scale field determinations that use the hydraulic
head/discharge analysis. This comparison could by useful in drainage pro
jects for which only direct measurements are available.
The average result obtained by the auger hole method (Table 8.7) was
lower than that derived from the hydraulic head discharge analysis:0.2 m/day
versus 0.6 m/day.
TABLE 8.7 Hydraulic conductivity auger hole method
No. of drain group
1 clay
2 clay/gravel
3 PVC/gravel
4 PVC
5 PVC/straw
6 PVC/esp.
7 PVC/coco
8 clay/gravel
Date
October '76 February '77 February '77
October '76 February '77
October '76 February '77
October '76 February '77 February '77
October '76 October '76 February '77 February '77
February '77 February '77
October '76 February '77 February '77
February '77
measured
Water
by the
table depth
m
1.15-0.80 -1.30 -
1.30 -1.15 -
1.10 -0.90 -
0.95 -0.40 -0.80 -
0.65 -1.20 -0.20 -0.55 -
0.45 -0.70 -
0.60 -0.50 -0.80 -
0.80 -
1.30 1.10 1.40
1.50 1.35
1.20 1.30
1.15 0.80 1.20
0.80 1.35 0.40 1.05
0.80 1.15
0.90 0.75 1.10
1.00
Depth of the hole
m
1.4 1.4 1.7
1.7 1.6
1.4 1.5
1.5 1.2 1.5
1.3 1.5 1.0 1.4
1.1 1.4
1.1 1.2 1.4
1.5 '
K
m/day
0.21 0.13 0.00
0.12 0.21
0.29 0.20
0.11 0.05 0.03
0.15 0.05 0.11 0.16
0.07 0.07
0.09 0.11 0.04
0.18
205
Table 8.7 shows that the K-values measured in f i e lds where the drains
function properly, are generally higher than those measured in f i e lds whose
dra ins have a high entrance res i s tance and where the water table seldom
drops below 1 m.
No sa t i s fac tory r e su l t s were obtained with the inversed auger hole
method used above the water t ab le , since hardly any f a l l of water level
was observed in the "pour-in holes". Augering in these s i l t y clay so i l s
may have sealed the so i l pores.
The r e su l t s obtained by laboratory measurements in undisturbed so i l
cores (Table 8.8) show a greater v a r i ab i l i t y and are even lower than
those measured by the auger hole method. Nevertheless, they reveal tha t the
s o i l i s an iso t ropic , since the hydraulic conductivity in v e r t i c a l d i r
ect ion i s always higher than that in horizontal d i r ec t ion , probably due
to the presence of root-holes and cracks. They also confirmed the decrease
of hydraulic conductivity with depth.
TABLE 8.8 Laboratory determinations of the hydraulic conductivity in undisturbed so i l cores
Depth Direction K
m m/day
0 - 0.50 horizontal 0.004 v e r t i c a l 0.970
0 - 0.30 hor izontal 0.0401
ve r t i c a l 0.690
0.50 - 0.85 horizontal 0.015 v e r t i c a l 0.730
0 . 8 5 - 1.00 hor izontal 0.001 v e r t i c a l 0.180
1 . 00 - 1.50 hor izontal 0.001 v e r t i c a l 0.010
1 . 5 0 - 2 . 0 0 hor izontal 0.001 v e r t i c a l 0.020
after subsoiling
206
The conclusion could thus be drawn that once the order of hydraulic
conductivity has been calculated from the experimental fields, the auger
hole method can be used to estimate the hydraulic conductivity at a large
number of sites for use in a later phase of large-scale drainage projects.
8.5 Determination of the entrance resistance
In the proximity of the drain the groundwater flow has to overcome
an extra resistance because the drain pipes are not pervious over their
entire surface; water enters only through the joints of the clay tile drains
or through the perforations of the plastic pipes.
The entrance resistance can be expressed by the following equation
(Engelund 1957, Cavelaars 1967):
h. h. W = — = -i (18)
e q sL
where
2 q = discharge per unit length of drain (m /day)
s = discharge per unit area (m/day)
L = drain spacing (m)
h. = difference in hydraulic head between the drain trench and the tile (m)
W = entrance resistance (day/m)
The entrance resistance is inversely proportional to the hydraulic
conductivity of the soil in the proximity of the drain, or
W = | (19) e K
207
where
K = hydraulic conductivity of the medium (m/day)
a = factor depending on the type of drain pipe
Equation (18) can be used to calculate the entrance resistance if
the head loss between the edge of the drain trench and the drain pipe, and
its corresponding discharge, are known. The difference in hydraulic head
was determined by two methods. The first consisted of a direct reading of
the piezometers installed in the drain trench and on the drain itself; the
second (indirect) method was by calculation once the position of the ground
water table was known.
8.5.1 Direct method of determining the W e value
The direct method is based on the correlation between the discharge
per unit length (sL) and the head loss between the edge of the drain trench
and the drain (h.).
by:
The W -value must be equal to the slope of the straight line expressed
h. = W (sL) (20) l e
The values of h. were plotted against the corresponding values of
sL and the line through the points was calculated by linear regression. The
slope of this straight line equals the entrance resistance W (Fig.8.21).
For each combination of drainage and filter materials, calculations
were made seasonally. The observation period commenced in October 1974.
For each period analysed at least 20 pairs of observations were compa
red. These observations pertain to periods with a falling water table.
A few determinations in which the correlation coefficient was less than
0.9 were ignored.
208
hj (mm)
200
180
160-
140
120
100-
80-
60
40-
20
0 10 20 30 40 50 60 70 80 90 sL(mm/day.mJ
Fig.8.21: Calculation of the W -value with the direct method.
The head loss near the drain (h.) was obtained from the difference
between the readings in the piezometer installed 0.20 m from the drain and
that on the drain itself (Fig.7.16).
An average value was taken for h. from two observations made in
piezometers placed at both sides of the drain. For drains with a length of
250 m an average value was adopted from readings in their two piezometer
lines.
The expression for the regression lines obtained reads as follows:
W sL + a e o
(21)
From a theoretical point of view, a should equal zero because the
hydraulic head should be zero when there is no discharge. Figure 8.21,
however, shows that, when the drain discharge becomes zero, there is still
a difference in hydraulic head between the piezometer 0.20 m from the drain
and the piezometer on the drain.
209
The existence of this residual value a may be due to two factors. One o
is that the drain is located in a less pervious layer, and the other is
that the piezometer 0.20 m from the drain was not placed inside the trench
but slightly outside it, thereby yielding a higher value for h. (Fig.8.22).
(real) i (observed) Fig.8.22: Difference between observed h. and real h..
The W -values for successive periods are presented in Table 8.9 and e
concern all the combinations of drainage and filter materials except PVC
pipes with straw cover (Drain Group 5 ) , which did not function properly
because the water flow inside the pipe was impeded.
Table 8.9 allows the conclusion that the entrance resistance remained
fairly constant during the period under analysis. Nevertheless, the increase
in entrance resistance (Fig.8.23) for the combination PVC/esparto shows
that the esparto filter gradually becomes less pervious. The data from the
PVC/cocos were somewhat irregular but show a similar tendency towards
increasing W -values.
These results indicate that the best combination for the soils of the
Alera field is clay pipes with gravel cover. These materials have the
lowest entrance resistance (W - 2 day/m) and the highest durability.
Corrugated PVC pipes with gravel cover can also be used although their
entrance resistance is more than twice that of the clay pipes with gravel
cover (W = 4 . 6 days/m). It must be taken into account that the exterior e
210
diameter of the p l a s t i c pipes i s 5 cm, whereas the diameter of the clay
t i l e s i s 8 cm. I t may well be that the entrance res i s tance of greater dia
meter PVC-pipe i s lower.
Table 8.9 Entrance r e s i s tance values (W , days/m)
Year
1974
1975
1976
1977
Season
Autumn
Winter Spring Summer Autumn
Winter Spring Summer Autumn
Winter Spring Summer
Mean entrance resistance (x)
Standard ion (s)
Number o (n)
deviat-
f periods
1
clay
8.4
9.0 3.6 5.0
---
3.6
-6.9 6.8
-
5.2
1.6
5
2
clay/gravel
1.2
1.2 3.7 0.8 1.2
0.71
1.2 0.8 1.7
3.2 0.61
1.41
1.5
1.0
12
No. of
3
PVC/gravel
6.9
10.4 2.3 6.7 3.91
6.2
-5.31
2.81
9.91
3.11
-
4.6
1.8
8
drain Group
4
PVC
13.4
11.0 12.2 12.8 8.11
12.2
-7.9
15.1
---
12.8
1.4
6
6
PVC/esparto
----
3.8
6.5 9.5 6.7
12.3
24.4 34.6 15.5
_2
_2
_2
7
PVC/cocos
----
12.1
23.7 21.0 22.3 19.7
35.3 35.3 15.7
21.7
1.7
4
8
clay/gravel
----
2.2
2.7 1.5 2.0 4.6
6.6 3.3 2.6
2.7
1.0
7
less than 20 pairs of observations
no average has been calculated since W increases with time (Fig. 8.23)
1975 h
Fig.8. 23: Increase of the W -value with time (PVC/esparto drains).
21!
Clay pipes without gravel have almost the same entrance resistance as
PVC pipes with gravel (W -5.2 days/m).
Corrugated PVC pipes without a filter show a high entrance resistance
(W - 13 days/m).
Plastic pipes with cocos and esparto filters show an even higher
entrance resistance than pipes without filters. The PVC/esparto drains
initially had a lower entrance resistance than the filterless plastic drains,
but the W -value gradually increased (Fig.8.23). Obviously there is no
advantage to be gained in using this envelope. Although the increase in the
W -value of PVC/cocos drains is not so clearly manifest, a similar tendency
exists.
Barley straw is not suitable as cover material since it rots easily
and clogs the pipe.These drains show very small but long-lasting discharges
and consequently the fall of the water table is slow. The drainage water
smells of H.S and forms a precipitate of sulphur at the drain outlets. Ob
viously, microbiological processes cause a rapid decomposition of the
straw envelope, at the same time reducing sulphates (abundantly present
in the groundwater) to H„S. At the outlets, sulphur bacteria oxidize H„S
to H-0 and S, forming whitish and sticky precipitates in the outlet pipes
(Fig.8.24).
The a -value of Eq.(21) was not constant for every drain group. Its variation during the observation period is shown in Table 8.10. A decrease in the a -value can be observed for those combinations that had
o
the lowest entrance resistance and showed a deeper and more frequent fall
of the water table, which can probably be explained by an improved permeab
ility in the drain trench.
212
Fig. 8. 24: Whitish sulphur precipitates in the outlet pipes of PVC/straw drains.
TABLE 8.10 Residual values obtained in the calculation of We (a in cm) o
N o . o f d r a i n g r o u p
Year
1974
1975
1976
1977
.
Season
Autumn
Winter Spring Summer Autumn
Winter Spring Summer Autumn
Winter Spring Summer
1
clay
36.6
32.4 34.6 30.0
---
23.8
-19.8 18.9
-
2
clay/gravel
20.8
21.7 14.5 18.3 14.0
3.61
6.9 3.6 4.5
0.0 6.11
9.41
3
PVC/gravel
30.8
25.5 26.2 24.7 18.81
6.0 -
28. 31
10.51
1.81
16.91
-
4
PVC
20.6
26.9 26.8 30.9 40.71
27.5
-23.01
27.7
24.21
--
6
PVC/esparto
----
50.0
35.2 34.8 32.4 33.5
24.7 10.8 22. 1
7
PVC/cocos
----
42.3
32.1 34.8 33.1 48.4
39.7 28.5 36.4
8
clay/gravel
----
27.3
23.1 20.5 19.6 22.1
18.2 19.2 22.3
less than 20 pairs of observations
213
8.5.2 Indirect method of determining the We value
The values of the entrance resistance obtained in the previous section
were checked by a second, indirect, method. The reason for doing so was
that errors may have been introduced into the calculations because of the
piezometers not being placed exactly at the edge of the drain trench. The
indirect method is based on the shape of the water table measured in the
piezometers installed midway between drains (10m), at 5 m, and at 1 m from
the drain (Fig.7.16). This analysis was originally developed by van Hoorn
(1960).
The discharge per unit length can be calculated by applying Darcy's
law:
q = - K g y (22)
The flow rate through a section of the region of flow at a distance
x from the drain is
q = s (0.5 L - x) (23)
where
2
q = discharge per unit length at one side (m /day)
s = discharge per unit area (m/day)
K = hydraulic conductivity (m/day)
y = hydraulic head at distance x from the drain (m)
x = distance variable (m)
L = drain spacing (m)
From Eqs. (22) and (23) the following differential equation is obtained:
K y | | = s (0.5 L - x) (24)
Integration of Eq. (24) leads to
K |- = s (0.5 L x - |-) + C (25)
214
For the boundary conditions x = 0.5L, y = h = hydraulic head midway
between the drains (Fig.8.25), the integration constant is:
C = | (K h2 - s (0.5 L)2)
a h2-y2cm2
8.000 -r (0,8m2)
7.000
6 000
5.000
4.000
3.000 •
2 .000
1.000 •
0 -0
^^\^-^~
25 ( 1 / 2 l - x ) 2 m 2
/ '
^^ •
.-—" B
5 '
81
/s,
^^H
^ ^ - - S 3
•— s 4
s 5
100
Fig.8.25: Calculation of the hydraulic head loss due to the entrance resistance by extrapolation of the groundwater table position.
Substituting into Eq.(25) gives:
K (y2 - h2) = - s {x2 + (0.5 L ) 2 - 2(0.5 L) x}
Solving for the discharge yields:
s K
1.2 2
h - y
(0.5 L - x):
(26)
215
Putting (0.5 L - x) equal to e and rearranging give s:
2 2
• ^ +y~= 1 (27) ^ 2 .2
which is the equation for an ellipse.
The semi-minor axis is h and the semi-major axis is h/K/s. If the
entrance resistance W = 0, the semi-maior axis is 0.5 L. Then e
L _ , |/K _ T 2 4 K h 2
2 h|/- or 1/ = (28)
Equation (28) is the Hooghoudt formula for steady flow above drain
1. If W > 0 the semi-major axis is long<
appears between the drains (h /K/s > 0.5 L).
level. If W > 0 the semi-major axis is longer and only part of an ellipse
For unsteady flow the Boussinesq equation also shows an almost elliptic
shape of the water table. This allows Eq.(26) to be used for unsteady flow.
From piezometer readings the y-values for distance x = 1 m and x = 5 m
are known. Likewise the h-value is known for x = 10m.
By plotting values of (h2 - y2) against those of (0.5 L - x) for
different s-values a set of straight lines can be drawn (Fig.8.25), each
having a slope of
tan a = |- (29) is.
From these straight lines the hydraulic head in the proximity of the
drain can be derived by extrapolation, since for x = 0, (0.5 L - x ) 2 = 100
and y = h. . ' l
For several values of h. and sL the regression line can be drawn, the
slope of which equals the entrance resistance (Eq.20).
With this indirect method the W -values for different combinations of e
drain and filter materials were calculated from the winter data of 1975
to 1977.
216
The results obtained are listed in Table 8.11, and can be compared
with the values derived by the direct method for the same period in which
h. was measured instead of calculated. The mean values already presented
in Table 8.9 are shown again in Table 8.11.
TABLE 8.11 Values of entrance resistance W in day/m obtained by two methods
2 8 3 4 6 7
No. of Drain group
clay/gravel clay/gravel PVC/gravel PVC PVC/esparto PVC/coco
For winter period
h. calculated
(Eq.26)
2.7 4.5 5.3
15.7 13.4 23.1
according to
h. measured
(Eq.20)
2.2 4.7
10.4 12.2 15.3 29.5
Mean value according to h. measured
(Tab.8.9)
1.5 2.7 4.6
12.8
-21.7
Table 8.11 shows that both methods give similar results, though deviat
ions occur in some periods, due possibly to errors in the observations.
The hydraulic conductivity can be derived from Eq.(29). Table 8.12
shows the K-values thus obtained for the periods in which W was calculated. e
TABLE 8.12 Calculated K-values
Period
Winter
Winter
Winter
Winter
Winter
Winter
Winter
Winter
Winter
Winter
Winter
Winter
'75
'77
'76
'77
'75
'75
'76
'77
'76
'77
•76
'77
2
8
3
4
6
7
No. of drain group
(clay/gravel)
(clay/gravel)
(PVC/gravel)
(PVC)
(PVC/esparto)
(PVC/coco)
Water table below surface level
m
>
> > > > > > > > > > >
0.4
0.8
0.4
0.5
0.4
0.2
0.5
0.4
0.2
0.2
0.4
0.4
K
m/day
0.61
1.70
0.67
0.80
0.61
0.36
0.36
0.67
0.58
0.52
0.55
1.46
217
If those K-values that deviate from the general tendency shown in
Table 8.12 are ignored, an average hydraulic conductivity of 0.6 m/day
is found (standard deviation s - 0.1). This value agrees fairly well with
that derived earlier for the clay tile drains with gravel cover.
From the analysis of those cases in which the hydraulic conductivity
differs greatly from the average value, the following conclusions can be
drawn:
1. When the groundwater table is deep (Drain Group 2 - winter 1977),
the calculated hydraulic conductivity is higher than the actual
one because some flow occurs below drain level.
2. When the entrance resistance is high (Drain Group 7 - winter 1977)
and the slope of the groundwater table is small (Fig.8.26), errors
of a few centimetres in the piezometer readings result in great
differences in the value of tan a and consequently in the value of
the hydraulic conductivity.
0.1-
0.2-
0.3-
0.4-
0<i-
0.6-
0 7 -
0.8-
0.9-
1.0-
1.1-
1.2-
X
22-Jan 1977
27Jan:l977
4- Feb.1977 \
I 2 - F e b . l 9 7 7 \
3-Mar.- l977 \
^
=10 m. £ i
22Jan. l977
1 27Jan. l977
1 / 4Feb. l977
' 12 Feb.1977
f 3-Mar. 1977
—
f
\ 5 x =
Fig. 8. 26: Shape of the water table where laterals have a high entrance resistance (Drain Group 7, PVC/cocos, winter 1977).
218
Nevertheless this method has its uses in determining the hydraulic
conductivity, especially in cases where the Boussinesq theory cannot be
applied because the drains have a considerable entrance resistance (Drain
Groups 3, 4, and 6).
219
9. Desalinization process
9.1 Stages of the process
The desalinization of the soils of the Alera experimental unit consisted
of two stages: an initial leaching period without crops, with intermittent
irrigation and drainage of the leaching water, followed by a second phase
in which moderately resistant to salt-tolerant crops were grown under irri
gation. During the second stage the percolation losses completed the desal
inization process.
The object of research in the first phase was to determine the leach
ing efficiency coefficients by comparing the actual desalinization process
with theoretical models. This would make it possible to predict the amount
of irrigation water required for the desalinization of soils with differing
initial salt contents and the duration of their leaching processes.
In the second phase the relation between crops and groundwater table
depth was studied for its use in assessing the drainage criteria to be
applied in further land reclamation projects. In addition some water manage
ment factors, e.g. field application efficiency, were derived.
The initial leaching of the area drained by Drain Groups 1 to 5 started
in September 1974 and continued until August 1975, after which the cropping
programme began. The leaching period of the area drained by Drain Groups
6 to 8 extended from autumn 1975 to summer 1976.
220
9.2 Water balance of the initial leaching phase
The water balance assessed during the first stage of the desalinization
process was made to determine experimentally the evaporation amount, and to
compare this value with those calculated by conventional methods. If they
agreed reasonably well, one could conclude that seepage and capillary rise,
which are other unknown components of the water balance, had been correctly
estimated.
-Dra
Fig.9.1: Water balance of the initial leaahing phase.
The water balance equation for the unsaturated zone reads:
I + P + G = E + R + A W (1)
where
I = effective amount of irrigation water (mm)
P = effective amount of precipitation (mm)
G = amount of capillary rise from the groundwater table (mm)
E = amount of evaporation (mm)
R = amount of deep percolation ("recharge"; mm)
AW = change in amount of moisture stored (mm)
If R is the net percolation (R* = R - G) Eq.(l) can be simplified
to:
221
I + P = E + R* + AW (2)
As the soil lacked natural drainage the water balance of the saturated
zone could be expressed by the following equation:
R* + S = Dr + pAh (3) a
where
R = net downward percolation (mm)
S = amount of seepage (mm)
Dr = amount of artificial drainage (mm)
\i = drainable pore space
Ah = change in water table height averaged over the entire drain spacing L (mm).
The water balance of the entire soil profile down to the impervious
layer, which coincides with the drain level, was obtained by Eqs.(2) and (3).
I + P = E + (Dr + yAh - S) + AW (4) a
During the initial leaching both precipitation and irrigation were
effective amounts because surface runoff did not occur.
According to the discharge hydrographs (Chap.8) both lateral seepage
and upward flow through the less pervious layer were negligible, thus S
equalled zero. Equation (4) then reduced to:
I + P = E + D r + yAh + AW (5) a
While the water table was falling, evaporation caused no significant
capillary rise. If it had, the hydraulic head hydrographs (Chap.8) would
have shown a steeper decline and greater differences would have been ob
served between the falls in winter and summer. Hence, the capillary rise
was negligible (G - 0) and deep percolation equalled net percolation
(R* = R).
222
If in the collector discharge hydrographs (Fig.9.2) periods of time
are chosen in which the discharge rate at the beginning and end of the
periods is equal, i.e. four or five days after an irrigation or heavy rain
fall, the depth of the water table should also be approximately equal.
S mm. day"
17T
Fig.9.2: Collector discharge hydrograph (irrigation).
Likewise, when the infiltration process is over, four or five days
after an irrigation the moisture content in the unsaturated zone should
correspond approximately to field capacity.
223
TABLES 9 . 1 a nd 9 . 2 : W a t e r b a l a n c e o f t h e i n i t i a l l e a c h i n g p h a s e
w i t h o u t c r o p s ( 1 9 7 4 - 1 9 7 5 ; 1 9 7 5 - 1 9 7 6 ) .
P e r i od I r r i g a t i o n R a i n f a l l D r a i nage Ac tua l Mean a c t u a l e v a p o r a t i o n e v a p o r a t i o n
mm rnm mm mm mm/day
1 9 7 4 - 1 9 7 5
Sept.8 - Oct. 8
Oct. 8 - Oct.28
Oct.28 - Nov.30
Nov.30 - Jan.19
Jan.19 - Feb. 8
Feb. 8 - Feb.15
Feb.15 - Mar.12
Mar.12 - Mar.22
Mar.22 - Apr.25
Apr.25 - May 12
May 12 - May 30
May 30 - Jun.19
Jun.19 - Jul.17
Jul.17 - Aug.15
Aug.15 - Aug.26
412 .2
120.0
119.9
128.2
--
126.1
-127.4
--
130.7
145.7
-110.2
87 .2
7 .7
45 .1
14.7
2 6 . 6
21 .9
10 .6
2 4 . 5
135.5
38 . 1
69 . 1
2 . 3
1.6
55 .4
11.5
-69 .7
107.8
75 .6
17 .5
12 .9
84 . 1
2 . 0
142. 1
8 . 9
11 .6
9 2 . 9
9 1 . 3
4 . 8
5 0 . 0
-58 .0
4 9 . 2
6 7 . 3
9 . 1
9 . 0
52 .6
22 . 5
120.8
29 .2
57 . 5
40 .1
56 .0
50 .6
71 .7
-2.9
1.5
1 . 3
0 . 5
1 . 3
2 .1
2 . 3
3 .2
2 . 3
3 .2
2 . 0
2 . 0
1.7
6 .5
1 9 7 5 1 9 7 6
Nov.19 - Dec.17
Dec.17 - Feb. 1
Feb. 1 - Feb.12
Feb.12 - Apr. 8
Apr. 8 - May 12
May 12 - June 2
June 2 - Jul.10
Jul.10 - Jul.13
Jul.13 - Jul.31
Jul.31 - Sept28
Sept28 - Oct.12
Oct.12 - Oct.24
-87 .2
-90 .5
85 .6
8 1 . 0
121.4
-119.6
110.3
--
7 0 . 3
8 .5
5 5 . 0
81 .1
4 2 . 7
2 2 . 0
77 .0
3 0 . 0
2 .7
104.8
5 8 . 0
33 .0
4 5 . 0
38 . 1
3 3 . 8
60 .2
65 .6
4 5 . 5
39 . 1
16 .0
4 5 . 0
44 .4
15.2
9 .7
2 5 . 3
57 .6
21 .2
111.4
62 .7
57 . 5
159 .3
14 .0
7 7 . 3
170.7
4 2 . 8
2 3 . 3
0 . 9
1.3
1.9
2 . 0
1 .8
2 .7
4 . 2
4 . 7
4 . 3
2 . 9
2 . 9
1.9
226
ru — E r— E <°
*? m w f - f - f -0) ffi o>
rO <\J —
F vi s,
Ml
<K S,
'— S" VI
•vJ -u « SH V) u « » w
t-~i
« s
^ « fc-" VI i .
<-K
x-VI
• * J
+J « !N VI
(
-Q t j
• * • »
• a - ^ 3 v» >« O +i
^s O a sr
rQ &H
!N
en s +* o « - 3 -w
a v\> SH
-e o +> n. 13
£ » O ft) A H->vO
^- a • ? i
K -W O K
• ^ Vs> -U +i <3 O s. a, o ct^ » 5)
<» T<1
« • 4 i
s • s» - ^ + i a o o Rj g
« t < l
- « « *« s,
• . <a Ml +i • « o» 3 • C35 V\>
• ^ «
fc, + i
. ^ s* « £ w U-i
Vfl VI
« 4-, SH
s CO
SH VM
-t-i
« 3
227
Some conclusions can be derived from a comparison of the curves of
Figs.9.4a and 9.4b. During the initial leaching, a precipitation deficit
occurred from June to September, due to the long intervals between consecut
ive waterings and the lack of rainfall. In this arid period there was no
capillary flow that could compensate the precipitation deficit. In fact,
once the surface soil became dry, evaporation ceased. Consequently, the
fall of the water table was scarcely influenced by evaporation.
Figure 9.4c shows the potential evaporation calculated by the Thornt-
hwaite method for the same period. The shape of the curve is similar to
that derived from the water balance (Fig.9.4a), but the calculated values
are lower than those determined experimentally, the difference being greater
in the cold months. The explanation for this difference is that the Thornth-
waite formula is based on mean temperature and heat index. It does not take
into account, as an evaporation factor, the dry winds which occur frequently
in the area during the cold months.
Figure 9.4d shows the evaporation from a free water surface calculated
by the Penman method and based on data from the Zaragoza observation station.
This station is representative of a climate similar to that of the Alera
field, but with a greater aridity (Chap.2). In addition, the value calculated
with the Penman method (E ) refers to a free water surface which has a dif-o
ferent roughness height and albedo than a moist soil.
To summarize, the evaporation values obtained from the water balance
agreed fairly well with those calculated by conventional methods, thereby
confirming that seepage and capillary rise were negligible.
228
9.3 Theoretical leaching approach
9.3.1 Series of reservoirs
The theoretical series-of-reservoirs model developed by van der Molen
(1973) divides the rootzone into four layers, each 25 cm thick, and regards
each layer as representing a reservoir.
Irrigation water supplied to the first reservoir percolates to the
second, from the second to the third, and from the third to the fourth
(Fig.9.5).
The volume of water in each reservoir corresponds with the moisture
content of each soil layer, and can be considered constant because the
water and salt movement occur at a moisture content near field capacity.
cm
25
50
75
100-
Ci
I _L_ c,
c2
,J_—1 I c
°1I
4 - - - L -c m
<L
Fig.9.5: Series-of-reservoirs mode I.
229
The salt balance in the first reservoir is expressed by the following
differential equation:
V d c c = (c. - c ) Q dt (7) fc 1 r
where
V = volume of the reservoir filled with water (1)
dcf = change in salt concentration of the soil solution at field capacity (meq/1)
c. = concentration of the influent (irrigation water; meq/1)
c = concentration of the effluent (meq/1)
Q = rate of flow through the system (1/s)
dt = differential interval (s)
T = V/Q = time of residence (s)
For the first reservoir the concentration of the effluent is
c = f c + (1 - f) ci (8)
where
f = leaching efficiency coefficient representing the percentage of irrigation water mixing with the soil solution
FromEqs.(7) and (8) the following differential equation is derived:
-^— - -fSdt (9) c, - c. V
f C 1
Integration of Eq.(9) for the boundary conditions t = 0, cf = c. and
t = t, c. = cT yields fc I
230
cT = c. + (c. - c.) e f t / T (10) I l 1 l
where
c = concentration of the effluent percolating from the first reservoir into the second (meq/1)
In a similar way the concentration of the effluents from the second,
third, and fourth reservoirs are obtained:
:II = Ci (cl ~ c i ) ~f e (c2 ~ C i ) e (11)
j . f \ f 2 t 2 -ft/T . , ft -ft/T cin = c i + ( c ! - ci> —r e + ( c2 - c i } -T e
+ (c3 - ci) e (12)
, . f3t3 -ft/T . , f2t2 -ft/T c = c. + (c - c.) e + (c, - c.) e +
IV i 1 i 6T3 2 i 2 T 2
s ft -ft/T , s -ft/T ,,,,. + (c3 - ci) — e + (c4 - ci) e (13)
The concentration values of Eqs. (10), (11), (12), and (13) can be ex
pressed either in meq/1 of soluble salts or in terms of electrical conduct
ivity, the latter being proportional to the former (Fig.9.6).
As the moisture content at field capacity is 40 per cent by volume, a
layer 25 cm thick will have a volume of water equivalent to 100 mm at field
capacity. If units of percolation flow of 100 mm are chosen, the value of
ft/T will be numerically equal to f.
231
Drainage water Op
Fig.9. 6: Correlation between a (meq/l) and EC (mmhos/am) in drainage water.
Using Eqs.(lO), (11), (12), and (13) and taking the average electrical
conductivity of the irrigation water constant at 0.4 mmhos/cm, theoretical
desalinization curves were obtained for different initial salt contents and
distinct leaching efficiency coefficients (Fig.9.7).
EC mmhos/cm.
2 5 - 50 cm.
5 0 - t 0 0 c m .
KX> 200 300 400 500 600 700 800 9 0 0 1,000 U&O 1,200 1,300 1(100 mm
Fig.9. 7a: Theoretical desalinization curves (Soil Site 2, f = 0.5).
232
EC mmhos/cm.^44 K
0 - 2 5 cm.
— — — 2 5 - 5 0 cm
- - - - - 5 0 - 1 0 0 cm.
TOO 200 300 400 500 600 700 80~0 90~0 [ 0 0 0 TTOO 1,200 [300 1400 r
Fig. 9. 7b: Theoretical desalinization curves (Soil Site 33 f = 0.5). ECjnmhos/cm.
TOO 200 300" 400 500 600 70~0 600 900 Ij300 [156 i^OO 13^0 1.400 rr
Fig.9.7c: Theoretical desalinization curves (Soil Site 33 f = 0.75).
233
9.3.2 Numerical method
If a constant leaching efficiency coefficient is taken, the series-of-
reservoirs model provides a semi-analytical solution to the leaching problem.
For soils such as those of the Alera field, however, where the degree of
cracking decreases with depth, a variable leaching efficiency coefficient
increasing with depth can be expected.
Introducing variable f-values leads to very complicated equations. For
this reason a simpler, numerical method is used.
If in the first layer a net amount (a) of irrigation water of concen
tration (c.) is mixed with an amount (b) of the soil solution that has an
initial concentration of (c ), the concentration of the percolation water
from the first layer (c ) to the second can be expressed as follows:
ac. + be CI ' a * b 0 4 )
The concentration of the water from the second layer (cTT) can be cal
culated in the same manner. If the amount of net percolation (a) is constant
throughout the soil profile, i.e. no moisture is retained, the expression
reads: ac + dc
C I I - a + d ( , 5 )
For layers of the same thickness and same moisture-holding capacity,
the amounts of moisture are the same and are equal to the moisture content
at field capacity (b = d = W,. ). re
Expressions similar to Eqs.(14) and (15) can be derived for the third
and fourth layers.
In Eqs. (14) and (15) a leaching efficiency coefficient equal to 1 is
assumed. In practice the net amount of percolation water can be adapted to
suit the possibly varying leaching efficiency coefficient of each soil
1ayer.
234
9.4 Actual desalinization process
To monitor the desalinization process, five soil sampling sites were
chosen in the area drained by Drain Groups 1 to 5 (Chap.7). One site was a
soil patch of high initial salinity (Site 3, EC = 45 mmhos/cm) and two
others had moderate salinity (Sites 2 and 5, EC = 1 0 mmhos/cm) and were
located in areas with different drainage conditions (Drain Group 2 clay/
gravel and Drain Group 5 PVC/straw). In the undrained control area Site 1
was chosen for a study of the salinity evolution under natural conditions.
Site 4 was located in a drained basin but leaching was caused only by rain
fall. In the area drained by Drain Groups 6 to 8 two extra monitoring sites
were selected, one of high initial salinity (Site 6, EC = 27 mmhos/cm) and
the other of moderate salinity (Site 7, EC = 7 mmhos/cm).
The leaching curve shows the relation between the decrease of the sal
inity, expressed in terms of electrical conductivity, and the average amount
of water drained by the laterals between which the monitoring site was loca
ted (Fig.9.8).
ECe mmhos cm" >7i
. 0 - 25cm.
<• 2 5 - 50cm.
° 50 - IOOcm.
200 300 400 500 600 700 BOO 9 0 0 WJOO U00 1,200 1,300 8-Nov. 7-Dec. 22-Feb. 4-Apr. 10-May. 4-Jul. 3-Sep. drainedmm. 22 Sep.
1974
Fig.9.8: Actual desalinization curve (Soil Site 2, moderate salinity).
235
Three desalinization curves were selected to show the leaching achieved
at three sites with differing initial salt content and drainage conditions.
0.200
0.175
0.150
0 125
0.100
0.075
0.050'
0.025'
0-
;;
i
\
%
\.
of dry soil
\
\ \ x 0 - 2 5 cm.
\ \ H 2 5 - 5 0 cm.
\ , *°. ° 5 0 - 1 0 0 cm
N
-0 100 200 300 400 500 600 700 8 0 0 900 1.000 1,100 1.200 drained
Sep 22 Nov.8 Dec7 Febr25 Apr.4 May (0 July4 Sep.3 m m
Fig.9.11: Actual leaching curve of a soil with moderate initial salinity (Soil Site 2) and good drainage conditions (Drain Group 1, clay/gravel).
Figure 9.11 shows the desalinization of a moderately saline soil (aver
age value of three layers: CI = 0.16%, EC =11 mmhos/cm), in which the
drainage system (clay tiles) functioned properly. After 850 mm of water had
been drained a soil profile of 100 cm was virtually desalinised (CI = 0.02%,
EC < 4 mmhos/cm). During the desalinization process no alkalinization was
observed since at the end of the leaching period (1,200 mm of drainage water)
soil pH remained around 8 and the SAR was equal to 6.
Figure 9.12 shows the leaching of a soil whose initial salinity was high
(CI = 0.80%, EC =43 mmhos/cm), and whose drainage system had a low entrance
238
resistance (PVC/gravel). Reliable results were only obtained for the top
two layers, so no desalinization curve has been drawn for the deeper layer.
In the upper 50 cm rapid desalinization was observed, 600 mm of leaching
water being sufficient to reduce the salinity level to below half of the
initial value (CI =0.3%, EC =18 mmhos/cm). After that, slower desalinization e
was observed and at the end of the initial leaching period (1200 mm of
drainage water) moderate salinity remained (CI =0.20%, EC =13 mmhos/cm),
or four times less than the initial value. No alkalinization symptoms were
observed during the leaching process and the pH-value remained near 8.
CI.% of dry soil
O 8
0 . 7
0 . 6
0 5
0 . 4
0 3
0 2"
o r
n-
.
^"v \
\ . + \ X * >• \ \
\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ * \ V
\ ' < ^
\
v^^
* 0 - 25 cm.
_—* 2 5 -50 cm.
0 Sep.22 1974
200 400 600 800 1,000 1,200 1,400 1,600 1,800 2.000 2200 2,400 drained n Nov8 Dec.7 Nov.25 Apr4 Moy.10 Jul.4 Sep.3
1975
Fig.9.12: Actual leaching curve of a soil with high initial salinity (Soil Site 3) and good drainage conditions (PVC/gravel).
Figure 9.13 shows the leaching of a moderately saline soil (CI =0.18%,
EC = 1 2 mmhos/cm), with poor drainage conditions caused by high entrance
resistance (PVC and PVC/straw). Here,the water table was rarely below 80 cm.
239
and causes scarcely any percolation to the water table.
241
Nevertheless, desalinization of the upper 50 cm was achieved and at the end
of the leaching period the soil was almost salt free (CI = 0.02%, EC = e
4 mmhos/cm). No reliable data were obtained for the layer below 50 cm.
CI % of dry soil 0 .17i
0.16
0.15
0.14
0.13
0.12
0.11
0.10
0.09
0.08
0.07'
0.06
0.05'
0.04
0.03
0.02
0.01
o
x 0 - 25 cm.
+—— 2 5 - 50cm.
- 5 0 - 1 0 0 c m .
S 0 I974
J F I975
Fig.9.15: Aatual leaching curve in a drained soil leached only with rain water.
No soil samples were taken to study the desalinization of soil layers
between a depth of 1 m and drain level, but to ascertain the variation of
the salt content in this deep soil layer, the drainage water was analysed
when the groundwater table was below 1 m.
For different initial salt contents and drainage conditions, the
Schoeller diagrams of Figs.9.16 and 9.17 show the decrease in salinity of
the drainage water during the leaching period. In the graph of Fig.9.18
the desalinization is expressed in terms of electrical conductivity.
242
- o ^ 0 0 ^ to *f>
K
G
Cs
in
«
02 Oj
Co
<» CO
« w
W)
'« w rS" +i
S, W
+i
« 3 K S O
fc • tn-N t<i
<5> 5 »
G> .« a s " (.•> r£f
to g M
s M)
« •< J
'« fc TO ^ r-J
TO ("i A " O
0 3
+i a
o \ 3>
- a K <o <a 0 +i K -O <^ « a. -W s
t- i 0 « s< to e> <s K
rC " ^ +> <3 , s S Q O "--
243
Cl.% of dry soil
0.150'
0.125-
o. ioo-
0.075
0.050
0.025
0-
— X — _
« —*—
f =0,25 f =0,50
actual desalinization curve
Fig. 9.19: Determination of the leaching efficiency coefficient (Soil Site 2, layer 0-25 cm).
CI % of dry soil
, f - 0 . 25
_ _ » f = 0 . 5 0
•••— • ac tual curve
100 200 300 400 500 600 700 800 900 1,000 1.100 1,200 drained mm
Fig.9.20: Determination of the leaching efficiency coefficient (Soil Site 2, layer 25 - 50 cm).
246
0.200
0.175
0.150
0.125-
0.100-
0.075-
0.050
0.025'
0 -
X&v -——— f = 0.50 ^fc\, — x — f = 0.70
^ \ f» 1.00
\ \ \ ° actual curve * \ N>
' \ \ \ \ \ \ \ \ \ \ \ \ -A N. • \ ' \ \
\ • • • • \ \
\ •-.\ V -. •• \ s \ •• \ N
\ l \ x \ -. \ X
*• \ ^ \ \ X
' '•• \ ^ •-. \ \ \ \ \ \ ' : . \ ^ -v
\ '••<NV \ \ -->v v.
100 200 300 400 500 600 700 800 900 1,000 1,100 1,200 drained mm.
Fig.9. 21: determination of the leaching efficiency coefficient (Soil Site 2, layer SO - 100 cm).
Two important conclusions emerged from the comparative study of the
leaching of a moderately saline soil (Soil Site 2). The first is that the
leaching efficiency coefficient is not constant but increases with depth.
The sec-ond is that the leaching efficiency coefficient is higher in the
first stage of desalinization and decreases gradually as the soil becomes
less saline.
The explanation for the first conclusion is that even though the soil
texture remains almost the same, cracks are less developed in the deeper
layers.
The second conclusion also has a logical explanation. At the beginning
of leaching the salts are accumulated at the aggregate surfaces, and are
easily washed out. At a later stage when the soil becomes less saline, some
salts remain in the interior of less pervious aggregates and the diffusion
of salts into the percolation water slows down.
247
The difference between the f-values obtained from Sites 3 and 2 may be
explained by Site 3 having a high initial salinity and being located in a
patch of rather compact soil with a lower infiltration rate than the rest
of the area. The amount of drainage water calculated previously corresponds
to the average amount of water percolating through the whole drainable area,
whereas the amount of water actually percolating at Site 3 was probably less.
Therefore, in reality, the leaching efficiency coefficient could be somewhat
higher than the calculated value.
At sites where the drains have a high entrance resistance, the actual
amount of leaching water may be greater than that measured in the drain
outlets, because some of the water may flow to adjacent areas drained by
laterals with a lower entrance resistance. If the real amount of leaching
water is greater than that used for the calculation,the leaching efficiency
coefficient should in reality be lower than the value calculated.
Notwithstanding these sources of error, the leaching efficiency coef
ficients reflect the differences in soil structure being higher in well-
structured soil patches like Sites 2 and 5. The values also reflect the
initial salt content since Sites 2 and 5, having a better soil structure,
had a lower initial salt content because of better natural leaching in the
past.
9.6 Prediction of the initial leaching requirements
To predict the initial leaching requirement, the numerical method was
used. For soils like those of the experimental field, an average leaching
efficiency coefficient of 0.5 for the upper soil layer (0 to 50 cm) was
assumed. For the deeper layer (> 50 cm) a value of 1.0 was used, provided
that drainage conditions were adequate.
Assuming a salinity corresponding to an EC of 15 mmhos/cm as being
representative of the saline soils of the fluvio-colluvial valleys, approx
imately 1 000 mm of percolation water must be drained by the field laterals
to leach the soil profile to a depth of 1 m (Fig.9.11).
252
If the desalinization is done by intermittent leaching (i.e.irrigation
water is applied when the water table is near drain level and the discharge
is approximately 0.5 mm/day), the average discharge during the leaching
period is 6 mm/day provided the entrance resistance of the laterals is low.
With intermittent leaching and an average irrigation cycle of 25 days,
about 6 months are required for the initial leaching. Taking into account
that heavy rainfall can occur during the leaching period and that the dis
charge will then be lower than with irrigation, the total duration of the
initial leaching period can be up to 8 months. Leaching should therefore
start in early autumn and continue to late May.
Although the initial leaching is performed during the months of low
evaporation, approximately 450 mm of evaporation must be added to the amount
of percolation water. The water supplied by rainfall varies considerably
from one year to another. Even so, the total irrigation requirement can be
estimated to vary between 1100 and 1400 mm, i.e. 8 to 10 applications of
140 mm each.
From the study of the initial salinity conditions and the behaviour
of the soils under leaching, it can be concluded that the initial salinity
correlates with soil physical properties that influence drainage conditions.
Important properties such as infiltration rate and permeability depend on
the compactness of the soil, irrespective of texture. The highest initial
salinities found in the experimental field corresponded to two soil patches
whose surface layers were compact. The amount of percolation water needed
to reclaim their upper layers (0 to 50 cm) is approximately 1300 mm (Fig.
9.27).
The main restraint to the reclamation of the severely saline soils
is the poor downward water movement through the soil profile. This, how
ever, can be improved by deep subsoiling and by local applications of
gypsum.
253
CI. °/o of dry soil
0 -25 cm
2 5 - 5 0 cm.
0 200 400 600 800 1.000 1,200 1,400 1,600 1.800 drained mm.
Fig.9.27: Prediction of teaching requirements to reclaim the surface layer (0-50 cm) of a highly saline soil (Soil Site 3).
9.7 Leaching requirement during the cropping stage
Once the initial leaching stage is over, the salinity level in the root-
zone corresponds to an EC of 3 iranhos/cm (CI < 0.02%), due to the presence
of slightly soluble salts. There is always a possibility, however, that
the soils may be resalinized by salts contained in the irrigation water and
salts transported by capillary rise from the groundwater.
To evaluate this hazard of secondary salinization the salt balance of
irrigated crops was obtained from the water balance for sugar beet, which
had been cultivated during the 1976 irrigation season and whose water balance
had been assessed to determine the irrigation efficiency and the percolation
losses (Table 9.3).
254
TABLE 9.3: Water balance for irrigated sugar beet (April - December 1976)
Period r i g a t i o n
nun
95 .8
41 .1
62 .1
72 .4
79 .4
89 .1
7 6 . 0
85 .4
83 .7
97 .9
105.7
83 . 5
90 .2
0
0
R a i n f a l l
mm
0
21 .7
0
14.0
61 .1
21 .9
40 .1
2 .7
0
0
4 5 . 8
11.0
28 .0
6 2 . 0
2 9 . 0
D r a i n
s u b s u r f a c e
mm
4 6 . 3
28 . 5
34 . 5
26 .9
6 0 . 3
27 .9
4 5 . 2
2 0 . 0
31 .9
30 .0
36 .4
34 .7
4 5 . 2
13.2
8 .9
a g e
s u r f a c e
mm
7 .0
2 . 0
1.8
1.8
0
0
1.5
1.6
0
0
0
0
0
0
0
Consumpt
mm
4 2 . 5
3 2 . 3
2 5 . 8
57 .7
8 0 . 2
83 .1
6 9 . 4
66 . 5
5 1 . 8
67 .9
115.1
5 9 . 8
7 3 . 0
4 8 . 8
20 .1
Lve u s e
mm/day
0
2 . 7
2 . 9
3 .4
4 . 2
4 . 6
6 . 3
6 .1
5 . 8
6 .2
6 .4
5 .5
4 .1
3 . 3
1.6
e a
0 .51
0 .26
0 .42
0 .60
0 .24
0 .69
0 . 39
0 .75
0 .62
0 .69
0 .66
0 .58
0 .50
--
Apr . 7 - Apr . 18
Apr . 18 - Apr . 30
Apr . 30 - May 9
May 9 - May 26
May 26 - June 14
J une 14 - J u l y 2
J u l y 2 - J u l y 13
J u l y 13 - J u l y 24
J u l y 24 - Aug. 2
Aug. 2 - Aug. 13
Aug. 13 - Aug. 31
Aug. 31 - S ep t . 11
S ep t . 11 - S e p t . 2 9
S e p t . 2 9 - O c t . 14
O c t . 14 - O c t . 29
This water balance was assessed by means of the equations and methodo
logy described in Section 2 of t h i s chapter. When data on discharge were not
ava i l ab le , percolat ion losses were calculated with the following equation:
R = uAh
An average value of Ah was determined from the readings of five piezometer
lines placed in the irrigated basin.
Figure 9.28a shows the variation in the consumptive use of sugar beet
throughout the irrigation season. The consumptive use was 894 mm, made up of
the net irrigation requirement, 557 mm, plus effective rainfall, 337 mm. For
an average field irrigation efficiency of approximately 0.53, the total irri
gation requirement was 1050 mm. In practice 1062 mm of water was applied.
255
Fig.9.28a: Consumptive use for sugar-beet derived from the water balance.
Fig.9. 28b: Potential evapotranspiration (Thornthwaite) for the same period as Fig.9.28a.
Assuming capillary rise to be negligible, the leaching requirement to
prevent secondary salinization by salts in the irrigation water could be
calculated with the salt equilibrium equation (van der Molen and van Hoorn,
1976).
256
(1 - f) EC + f EC.
R*= ( E - F ) f (ECf -EC.) ' ( 2 A )
fc 1 where
R = net leaching requirement (mm)
E-P = precipitation deficit (mm)
f = leaching efficiency coefficient
EC. = electrical conductivity of irrigation water (mmhos/cm)
EC. = electrical conductivity of the soil solution at field capacity (mmhos/cm)
The following assumptions were made: average value of f is 0.75 for a
depth to 1 m, EC. =0.4 mmhos/cm, ECf = 6 mmhos/cm, and the precipitation
deficit E - P = 557 mm. Substituting these values into Eq.(24) yielded the
leaching requirement: R = 239 mm.
Hence, there is no risk of secondary salinization by the irrigation
water if conditions of good drainage are maintained, because the measured
percolation losses (490 mm) are twice the leaching requirement. In fact,
during the irrigation season the salt content of Soil Site 4, expressed in
terms of chloride percentage, remained below 0.02 (equivalent EC = 3 to
3.5 mmhos/cm).
Capillary rise from the water table is indeed negligible because seepage
from adjacent slopes is caught by interceptor drains. Some capillary rise
may occur in the fallow period after winter cereal growing if seepage from
adjacent irrigated summer crops recharges the groundwater. Even so, any
salts accumulated in the rootzone during the fallow period can be leached
by one or two irrigations applied after the cereal harvest and just before
the seed bed is prepared for the next crop.
To summarize, secondary salinization after reclamation can be prevented
if two aspects of water management are borne in mind: i.e. good drainage
conditions should be maintained by keeping the field laterals and collector
drains clean, and irrigated crop rotation of lucerne, maize, and sugar beet
should be practised.
257
10. Subsurface drainage system
10.1 Drainage criteria for unsteady-state conditions
One of the objectives of the research undertaken in the experimental
fields was to assess drainage criteria that could be used in large-scale
drainage and reclamation projects. To meet this objective, the following
steps were taken:
Study of the relation between crop yields and water table depth for
irrigated summer crops and for winter cereal crops irrigated
only in spring.
Study of the relation, especially in winter, between water table
depth and the mobility of agricultural machinery on the land.
Study of the relation between heavy precipitation and the
corresponding rise in water table.
Study of the relation between percolation losses from irrigation
and water table rise.
After these studies had been completed, drainage criteria for unsteady
flow were derived to meet both the requirements of crops and the mobility
of agricultural machinery.
The next step was to convert the unsteady-state criteria to steady-state
criteria, these being easier to use in drainage projects.
Finally these criteria were compared with those assessed for the ini-
258
tial leaching requirements (Chap.9). This allowed the duration of the desal-
inization period to be predicted because the drainage system was designed
only to meet the requirements of irrigated agriculture.
10.1.1 Relation between crop yields and water table depth
To study the relation between crop yields and water table depth, data
from the Alera experimental field were used. The data covered the yields of
cereals and vetch grown during two consecutive winters (Figs.10.1, 10.2 and
10.3) and the yields of irrigated sugar beet, maize, and lucerne grown
during two consecutive summers.
Fig.10.1: View of wheat just before the heavy rainfalls of winter 1977.
259
TABLE 10.2: Crop yields in kg/ha and relative yield
N o . o f D r a i n g r o u p
Wheat
Barley
1
clay
5,250
0.95
5,500
1.00
2
clay/gravel
5,250
0.95
5,500
1.00
3
PVC/gravel
5,600
1.00
•
4
PVC
4,570
0.82
•
5
PVC/straw
4,570
0.82
4,430
0.81
* Crop yield affected by soil salinity
A comparison of Tables 10.1 and 10.2 shows that if the water table re
mained within the upper 50 cm for 3 consecutive days, no obvious harm was
done to winter cereals. This was even true if the 3-day period coincides
with the time of the crop's maximum development (i.e. tillering and ear
formation stages). This conclusion could be drawn because the 5 per cent
difference in yield between Drain Group 3 and Drain Groups 1 and 2 is not
significant. There is, however, some evidence that if the water table re
mained within a depth of 50 cm for 9 or more consecutive days, the yield
for both wheat and barley decreased from 15 to 20 per cent.These reductions
were observed in the 1977 winter season notwithstanding a late-winter nitro
gen application. Ammonium nitrate (33.5% N) at 200 kg/ha was applied when
crops in areas with shallow water tables showed nitrogen deficiency symptoms.
The yield reductions agree with the results obtained by SIEBEN (1974) in the
IJssel Lake Polders. Sieben observed a reduced supply of nitrogen from the
soil due to a slowing down of mineralization as a result of shallow winter
water tables. In his experimental plots he observed that this reduction in
the nitrogen supply could be compensated by an extra dressing of nitrogen.
An estimate of the relation between sugar beet yields and water table
depth could be made by comparing the data of Tables 10.3 and 10.4. It seems
that a period with a water table within a depth of 50 cm, for 3 consecutive
days is not critical. This water table proved to be critical at the seedling
stage in poorly drained plots (PVC/straw drains) making transplantation ne
cessary at the singling stage.
262
TABLE 10.3: Number of consecutive days in which the water table was within the depth indicated in cm (irrigation season 1976)
N o o f D r a i n g r o u p
2
clay/gr
25 50
avel
100
cm depth
3
- -3
1 6
- -7
15
8
17
10
12
10
3
PVC/gravel
25 50 100
cm
1
-1
1
--
depth
6
-3
3
4
9
15
21
21
20
19
15
25
cm
------
4
PVC
50 100
depth
8
3
--1
8
15
31
27
31
29
30
5
PVC/straw
25
cm
1
-----
50 100
depth
9 30
5 31
- 30
1 31
3 31
8 30
April
May
June
July
August
September
TABLE 10.4: Sugar beet yield (in kg/ha) and relative yield (1976)
N o . o f D r a i n g r o u p
2 3 4 5
clay/gravel PVC/gravel PVC PVC/straw
Yield^ (kg/ha) 45,500 50,500 40,400
Relative yield 0.90 1.00 0.80
45,500
0.90
* Yields were only estimated because of difficulties in harvesting
263
These conclusions were confirmed during the 1977 irrigation season.
A comparison of the data of Tables 10.5 and 10.6 shows that sugar beet grows
well (Fig.10.4) if the water table is kept between 75 and 100 cm and, if
it rises above this level, does not remain within the upper 50 cm of the
soil for longer than 3 to 4 days.
Fig.10. 4: View of sugar-beet grown during the first irrigation season after the desalinization period.
TABLE 10.5: Number of consecutive days in which the water table was within the depth indicated (irrigation season 1977)
June
July
August
September
N o . o f
6
PVC/esparto
25 50 75 100
cm depth
1 6 20 50
1 2 11 31
1 2 8 31
1 2 5 30
D r a i n g r
7
PVC/coco
25 50 75 100
cm depth
4 20 30
2 11 31
I 4 30
1 3 19
o u p
8
clay/gravel
25 50 75 100
cm depth
3
2
1 2
1
6 24
3 31
4 22
2 18
264
TABLE 10.6: Crop yields and relative yields (1977)
Lucerne* (kg/ha)
Sugar beet (kg/ha)
N o .
6
PVC/esparto
1,235
0.57
42,500
1.00
o f D r a i n
7
PVC/coco
1,685
0.78
42,500
1.00
8 r o u p
8
clay/gravel
2,160
1.00
42,500
1.00
Two outs of lucerne sown in spring 1977
Lucerne is more sensitive to high water tables (Table 10.6, Fig.10.5).
Table 10.8 gives the hay yields obtained from lucerne sown the previous
autumn in plots with different drainage conditions (Table 10.7).
Fig.10. 5: Differences in drainage conditions after the heavy rainfalls of winter 1977 are obvious in a first year lucerne.
265
TABLE 10
June
July
August
7:
September
Numb with
25
4
2
2
2
er of consecutive days in which the in the depth indicated (irrigation
2
clay/gravel
50 75 100
cm depth
5 6
3 4
4 5
4 5
20
10
16
7
N o . o f
3
D
PVC/gravel
25 50 75
cm depth
5 6 10
2 3 10
3 6 10
3 4 8
100
30
31
31
23
r a
water table was season 1977)
i n
25
5
1
2
3
g r o u p
4
PVC
50 75 100
cm depth
9 22 30
10 25 31
14 28 31
8 17 30
25
5
1
3
3
PVC/
50
5
straw
75 100
cm depth
20
19
24
8
30 30
31 31
30 31
14 30
TABLE 1 0 . 8 : Lucerne y i e l d and r e l a t i v e y i e l d (1977)
N o . o f D r a i n g r o u p
c l a y / g r a v e l PVC/gravel
4
PVC PVC/straw
Y i e l d ( kg /ha )
R e l a t i v e y i e l d
12,195
1.00
7 ,600
0 .62
5,780
0.47
5,415
0.44
Five outs of first year lueevne, sown in autumn 197'6
A good yield was obtained where the water table remained within the
surface 25 cm for less than 3 days, within 50 cm for 4 to 5 days,and within
75 cm for 5 to 6 days (Fig.10.6).A s igni f icant yield decrease of 40 per cent
occurred where the water table remained for 10 consecutive days within a
depth of 75 cm and continuously within 100 cm. These r e su l t s were confirmed
by the y ie lds of lucerne sown in spring 1977 (Table 10.6). The r e su l t s ob
tained for maize in 1977 (Tab.10.9) are s imilar to those for lucerne (Tab.
10.8).
TABLE 10.9: Maize y ield and r e l a t i v e y ield (1977)
N o . o f D r a i n g r o u p
c lay/gravel PVC/gravel
4
PVC PVC/straw
Yield (kg/ha) 5,800
Relat ive y ield 1.00
4,000
0.69
1,730
0.30
1,180
0.20
266
~ " a * - . ' " " " '
Fig.10.6: Detail of the growth of the lucerne of Fig.10. 5 in well-drained areas.
10.1.2 Relation between water table depth and mobility of agri
cultural machinery
The winter of 1976-1977 being wetter than usual, the water table
remained near the soil surface in the worst drained areas. These conditions
affected the seedbed operations for winter wheat.
Table 10.10 shows water table depths in three areas with different
drainage conditions and their effect on the mobility of agricultural
machinery for seedbed preparation (Fig.10.7).
267
lifl.l'i.
Fig.10.7: Differences in soil colour show distinct drainage conditions affecting mobility of agricultural machinery.
TABLE 10.10: Relation between water table depth and mobility of agricultural machinery (winter 1976-1977)
N o o f D r a i n g r o u p
PVC/esparto
25 50 75 100
cm depth
PVC/coco
25 50 75 100
cm/depth
clay/gravel
25 50 75 100
cm depth
No.of days with water table within the indicated depth
Possibility of seedbed preparation
3 16 29 31
negative
3 13 25 31
negative positive
From Table 10.10 it can be concluded that if the water table remains
almost continuously within 75 cm, the mobility of agricultural machinery
is impeded.
268
10.1.3 Rainfall distribution and water table rise
Data on maximum rainfall during 1 to 6 consecutive days expected for
different return periods are shown in Table 10.11. These figures were de
rived from the depth duration frequency curves discussed in Chap.2.
TABLE 10.11: Maximum rainfall (in mm) distribution (Gumbel analysis)
Return (yea
1
2
5
10
20
100
per r s )
LOd 1
14.5
40.0
55.0
67.5
79.0
101.0
C o n s e c u t
2
24.5
54.0
72.0
82.0
97.0
121.0
3
29.0
58.5
76.5
88.0
100.0
124.5
. v e d
4
33.0
62.5
79.0
92.0
103.0
127.0
a y s
5
36.0
65.5
82.5
95.0
106.5
131.5
6
38.0
67.5
85.5
98.0
109.5
135.0
The probabi l i ty of long duration maximum r a i n f a l l i s low since the pre
c i p i t a t i on to be expected in 4 to 6 consecutive days i s scarcely more than
t ha t expected in 3 days, as shown in Table 10.11.
The Gumbel p robabi l i ty analysis was applied separately for the i r r i g a
t ion season from March to October. Table 10.12 gives the r e su l t s obtained.
TABLE 10.12: Rainfall in mm during 3 consecutive days expected in the i r r i g a t i on season
Return period Rainfal l
(years) (mm)
1 11.0
2 50.0
5 68.5
10 81.5
20 93.0
100 120.0
269
A comparison of Tables 10.11 and 10.12 shows that there is no marked
seasonal distribution of heavy rainfall, although the rainfall of the
autumn-winter season is slightly higher than that of spring-summer. Hence
the annual values shown in Table 10.11 could be used to assess the winter
drainage criteria.
Assuming that the water table is near drain level and that the moisture
content of the unsaturated zone corresponds approximately to field capacity,
it can be expected that the water table will rise to the soil surface at
least once every 2 years. In fact, if the effective porosity is 4 per cent,
the storage capacity of the soil profile above the average drain level
(1.20 m) is approximately 50 mm. Although it is difficult to assess the
soil moisture content before rainfall, it can be expected that irrigated
soils will have a moisture content near field capacity during the autumn-
winter season.
10.1.4 Relation between percolation losses from irrigation and water table rise
Observations made during two irrigation seasons (1976 and 1977) showed
that water applications varied from 80 to 90 mm in the larger basins (300 x
50 m) where the slope is almost zero, while in smaller basins (180 x 50 m)
where the slope of the furrows is 0.1 per cent the water application was
about 60 mm.
The interval between two consecutive applications varied from 12 to
14 days during the summer season. In the smaller basins it was generally
1 or 2 days shorter.
The irrigation practice was similar for sugar beet, lucerne, and maize,
except that sugar beet received one or two water applications more than
lucerne in April, and maize was not irrigated in May.
Water losses were almost solely due to percolation, since little "tail
water" (no more than a few mm) was left after an irrigation.
270
The application efficiency was lower in the larger irrigation basins (e =
0.55 to 0.60) than in the smaller ones (e = 0.70 to 0.75).
During the period of maximum water requirement, i.e. from mid-July to
mid-August, the amount of water applied was about 90 mm every 11 days. The
irrigation efficiency, however, was greater because the high evapo-trans-
piration (5 to 6 mm/day) increased the soil moisture deficit.
The water lost by percolation varied from 30 to 35 mm in the larger
basins and from 15 to 20 mm in the smaller ones. If the entrance resistance
of the drainage material was low, part of this water flowed directly into
the drain trench. The amount of water recharging the groundwater was thus
estimated at 25 to 30 mm for the larger basins. It caused the water table
to rise by about 0.7 m (Fig.8.4).
10.1.5 Assessment of drainage criteria for unsteady-state conditions
Two periods were distinguished: the summer irrigation season and the
winter season with spring irrigation.
Winter season
In the previous sections it was established that in winter the water
table can remain within 50 cm for about 3 days without any appreciable re
duction in the yield of cereals. However,if the water-logging lasted 9 days
the yield was reduced by 15 to 20 per cent, which is a considerable loss.
No harm was done if the water table remained within 25 cm for 1 or 2 days.
Soil conditions at the start of seed-bed preparation must be such that
the movement of agricultural machinery is not impeded. Assuming that the
critical water table depth to ensure such conditions is 65 cm and that
seed-bed preparation should not be delayed longer than one week, the winter
drainage criteria for unsteady flow as given in Table 10.13 could be applied.
271
TABLE 10.13: Winter drainage criteria for unsteady flow*
Drawdown of water table (from - to)
(m)
0.00 - 0.25
0.25 - 0.50
0.50 - 0.65
h o
(m)
1.20
0.95
0.70
h t
(m)
0.95
0.70
0.55
t
(days)
2
2
4
Average drain depth = 1.20 m; y = 0.04
Summer season
If irrigation water is applied every 12 days and each time causes the
water table to rise an average of 0.7 m, a gradual rise of the water table
will occur unless the drainage system is capable of lowering the water table
by 0.7 m during each irrigation interval.
To maintain an optimum moisture content of the rootzone during the
irrigation season, the depth to the (fresh) groundwater should not be more
than 1 m.
If the water table is approximately 0.8 m deep at the start of an
irrigation, the maximum hydraulic head at the start of water table draw
down is:
h = h + - = 0.A + 0.7 o t u
1.1m (1)
As irrigated crops seem to tolerate a water table within a depth of
0.5m for at least 3 days, the drainage criteria applicable during the
irrigation season for unsteady flow were tentatively assessed as shown in
Table 10.14.
272
TABLE 10.14: Unsteady-state drainage criteria for the summer season*
Drawdown of water table (from - to)
(m)
0.10 - 0.25
0.25 - 0.50
0.50 - 0.70
0.70 - 0.80
h o
(m)
1.10
0.95
0.70
0.50
ht
(m)
0.95
0.70
0.50
0.40
t
(days)
1
2
4
5
^Average drain depth is 1.20; y = 0.04
10.2 Drain spacing
With the use of the unsteady-state drainage criteria and the hydrolog-
ical factors experimentally derived in Chap.8, the drain spacing was calcu
lated with the Boussinesq equation:
where
L
K
h o
ht
t
u
. 4.46 K h h t . 2 _ o t
u(hQ - ht)
drain spacing (m)
hydraulic conductivity (m/day)
hydraulic head midway between drains immediately after instantaneous recharge (m)
hydraulic head midway between drains after t days (m)
time (days)
drainable pore space (dimensionless)
(2)
Table 10.15 shows the drain spacings calculated with Eq.(2), using
the criteria for both the winter and the summer season.
273
TABLE 10.15: Drain spacing computed by Eq.(2)*
Period
Winter
Summer
Drainage criteria
h o
(m)
1.20
0.95
0.70
1.10
0.95
0.70
0.50
h t
(m)
0.95
0.70
0.55
0.95
0.70
0.50
0.40
t
(day)
2
2
4
1
2
4
5
Hydrological
K
(m/day)
1.0
1.0
0.6
1.0
1.0
0.6
0.6
factors
V
0.04
0.04
0.04
0.04
0.04
0.04
0.04
Drain spacing
m
32
24
26
28
24
22
26
Average drain depth = 1.20 m
An average drain depth of 1.20 m was assumed. This means that for la
terals with a length of 200 m and a slope of 0.1 per cent, the initial drain
depth is 1.1 m and the outlet to the collector drains is at 1.3 m. For the
lateral drains to meet the drainage requirements of both the summer and
winter season, they must be installed at a depth of 1.20 m and spaced at 25 m.
10.3 Drainage criteria for steady-state conditions
In the above calculation of drain spacing with unsteady-state criteria,
the entrance resistance was not taken into account. In Chap.8, however, it
was shown that the entrance resistance of some drainage and filter materials
is high. Hence it was necessary to take the entrance resistance into account
in calculating drain spacings. For this purpose, steady-state drainage
equations with criteria corresponding to those defined earlier for unsteady-
state conditions were used.
The steady-state criteria were derived from the Hooghoudt equation
for drains located on an impervious layer:
4 K h (3)
274
Introducing into Eq.(3) the mean h-values and the i r corresponding drain
spacings and hydraulic conductivity values taken from Table 10.15 yielded the
corresponding values of s. In t h i s way the s teady-s ta te c r i t e r i a equivalent
to those for the actual unsteady-state conditions were derived (Table 10.16).
TABLE 10 .16 : Conver s ion of d r a i n a g e c r i t e r i a
U n s t e a d y - s t a t e c o n d i t i o n s S t e a d y - s t a t e c o n d i t i o n s
P e r i od h o
(m)
t
(m) ( days )
a v e r age wa t e r t a b l e d ep t h
(m) (m) (mm/day)
Win t e r
Summer
1.20
0 . 95
0 .70
0 . 95
0 . 70
0 . 55
1.10 0 . 95
0 . 95 0 .70
0 .70 0 . 50
0.50 0.40
0.1
0 . 4
0 . 6
0 . 2
0 . 4
0 . 6
0 . 7
1.1
0 .8
0 . 6
1.0
0 .8
0 .6
0 . 5
4 . 7
4 . 4
1.3
5.1
4 . 4
1.8
0 . 9
*Ave r age d r a i n d ep t h = 1.2 m; p = 0 . 04
From Table 10.16 an average s/h r e l a t ion for s teady-s ta te conditions -3
of 3.9 x 10 was obtained. Table 10.17 gives values of s, h, and water
t ab le depth corresponding to th i s s/h r e l a t i on .
TABLE 10 . 17 : D r a i nage c r i t e r i a f o r s t e a d y - s t a t e c o n d i t i o n s *
Water t a b l e d ep th
(m)
0 . 25
0 . 50
0 . 75
1.00
h
(m)
0 .95
0 .70
0 . 45
0 . 20
s
(mm/day)
3 .7
2 .7
1.8
0 . 8
Average d r a i n d ep t h = 1.2 m; y = 0 . 04
275
A water table depth of 0.5 m under steady-state conditions means that
the hydraulic head midway between drains is 0.7 m and the discharge 2.7
mm/day, or 33 mm in 12 days, which agrees with the drainage requirements
in the summer season.
If in the initial leaching period the average drain discharge is
3 mm/day, approximately 11 months would be required to drain 1 000 mm of
leaching water, which seems reasonable in view of the leaching requirements
calculated in Chap.9. Hence, the criteria developed for the cropping period
also yield satisfactory drainage conditions for the desalinization process.
10.4 Drain spacings required for different combinations of drainage and filter materials
10.4.1 Steady flow
Since the entrance resistance must be taken into account, the Ernst
equation (Ernst, 1962) could be used to calculate drain spacings. According
to the Ernst theory for steady flow, the hydraulic head can be divided into
three components corresponding to vertical, horizontal, and radial flow.
To these a fourth component, corresponding to the entrance flow, was added:
h + h, + h + h. (4) v n r l
The general expression of the Ernst equation is:
D 2 h " S IT + S FKD + S L Wr + S L We (5)
v where
h = hydraulic head midway between drains (m)
s = discharge (m/day)
276
D = average thickness with regard to vertical flow (m) v
K = hydraulic conductivity for vertical flow (m/day) v
L = drain spacing (m)
2 KD = transmissivity (m /day)
W = radial resistance (day/m) r
W = entrance resistance (day/m)
In the soils of the Alera experimental field the vertical and radial
flow components are negligible, the flow being restricted to the profile
above drain level. Equation (5) thus reads:
h - \ + h i • s ra+ s L W e ( 6 )
For drains located on an impervious layer, D equals h/2 and Eq.(6)
changes to
h " S Ah + S L We (7)
Equation (7) was used to calculate drain spacings for different combin
ations of drainage and f i l ter materials. For this calculation the drainage
cr i ter ia for steady flow assessed in the previous section and the entrance
resistances calculated in Chap.8 were used. Table 10.18 shows the results .
TABLE 10.18: Relation between entrance resistance and drain spacing according to Eq.(7)
W (day/m)
L (m)
Drain density (m/ha)
1
clay
5.2
16.0
625.0
N o .
2,8
clay/gravel
2.1
18.0
555.5
o f D r
3
PVC/gravel
4.6
16.5
606.0
a i n g r
4
PVC
12.8
12.0
833.0
o u p
6
PVC/esparto
16.0
10.5
952.5
7
PVC/cocos
21.7
8.5
1176.5
277
l n T T = - - h r = - a t (.3) o e
with ot = 1/y L W , known as the reaction factor. e
Then again it holds that
h = h e"at (14) t o
Solving Eq.(13) for L yields a formula that enables drain spacings to
be calculated for unsteady flow through a high resistance:
L u W ln(h /h ) ° 5 )
e o t
Table 10.20 shows t he r e s u l t s ob t a ined wi th Eq . ( 15 ) .
TABLE 10 . 20 : R e l a t i o n be tween e n t r a n c e r e s i s t a n c e and d r a i n s p a c i ng a c c o r d i n g t o Eq . ( 15 )
Type of D ra inage c r i t e r i a d r a i n s W 7 T L Remarks
e h h t o t
(day/m) (m) (m) (day) (m)
PVC 12 .8 1.20 0 . 95 2 16.7
0 . 95 0 .70 2 12.8 c r i t i c a l
PVC/ e s p a r t o 16.0 1.20 0 .95 2 13.4
0 . 9 5 0 . 70 2 10 .2 c r i t i c a l
0 .70 0 .55 4 25 .9
PVC/coco 21 .7 1.20 0 . 95 2 9 .9
0 . 95 0 .70 2 7 .6 c r i t i c a l
0 .70 0 . 55 4 19.1
280
The results obtained from the different calculation methods (Tables
10.18 to 10.20) show a good agreement. They all reveal that if drainage
and filter materials of high entrance resistance are used, the drain density
(m/ha) is twice that needed with drains of low entrance resistance. Moreover,
materials with a high entrance resistance involve more risk of failure than
good materials that allow a wider spacing. If the entrance resistance of
some drains is double the average amount, such drains would be a practical
failure. If there were a tendency for W to increase with time, as is pro
bably so with PVC/esparto drains, the situation would be even worse.
In choosing drainage and filter materials a decision in favour of low
entrance resistance drains is the obvious one, even if the costs are higher,
since these will be offset by a lower drain density.
281
List of symbols
SYMBOL DESCRIPTION DIMENSION
A
a
B
b
D
d
D'
Dr
a
E
EC
ESP
e
e a
f
G
h
cross-sectional area (m )
geometric factor, constant
drain length (m)
bottom width of channel (m) constant
salt concentration (meq/litre) correction for drain spacing (m)
hydraulic resistance of semipervious layer for vertical flow (day)
thickness of aquifer or saturated layer (m) thickness of layer below drain level (m)
thickness of equivalent depth in Hooghoudt equation; inside diameter of pipe drain (m)
thickness of layer through which vertical flow occurs (m)
amount of artificial drainage (mm) rate of artificial drainage (mm/day)
amount of evapotranspiration (mm) rate of evapotranspiration (mm/day)
electrical conductivity (mmhos/cm)
exchangeable sodium percentage
base of natural (Napierian) logarithm
field irrigation efficiency
leaching efficiency coefficient in regard to irrigation water
amount of capillary rise of groundwater (m) rate of capillary rise (mm/day)
hydraulic head; height of water table above drain level midway between drains (m)
component of the hydraulic head due to horizontal flow (m)
dimensionless
L
L dimensionless
L
L L
L
£ , -I T -ohm cm
dimensionless
dimensionless
dimensionless
dimensionless
L LT
•1
282
SYMBOL DESCRIPTION DIMENSION
h. component of the hydraulic head due to entrance flow (m)
h height of water table above drain level midway between drains after instantaneous recharge (m)
h component of the hydraulic head due to radial flow (m)
h height of water table above drain level midway between drains at any time t (m)
h component of the hydraulic head due to vertical flow (m)
Ah average fall of water table between drains (m)
I effective amount of irrigation water (mm)
I, basic infiltration rate (mm/day) bas
i hydraulic gradient
j reservoir coefficient (day)
K hydraulic conductivity (m/day)
K' hydraulic conductivity for vertical flow (m/day)
K hydraulic conductivity of the layer above drain level (m/day)
1L hydraulic conductivity of the layer below drain level (m/day)
2 KD transmissivity of water-bearing layer (m /day)
L drain spacing (m)
L drain spacing for horizontal flow (m)
n number; Manning roughness coefficient
P effective amount of precipitation (mm) precipitation (mm/day)
pF logarithm of the water tension in cm water column
3 Q discharge (m /s)
rate of flow through a system (1/s) 2
q discharge per unit length (m /day)
R amount of deep percolation (mm) percolation rate (mm/day) hydraulic radius (m)
R amount of net downward percolation leaching requirement (mm)
net percolation rate (mm/day)
L
L
L -1
LT
dimensionless
T
LT
LT
LT
LT"1
LV L
L
dimensionless
L LT
•1
dimensionless
LV
LV1
L T -L
LT
283
SYMBOL DESCRIPTION DIMENSION
RSC
S
SAR
s
t
u
V
V
V z
W , f c
W e
W r
x, y,
X
y
a
<5
A
M
radius (m) correlation coefficient
residual sodium carbonate value (meq/1)
amount of seepage (mm) seepage (mm/day)
sodium adsorption ratio
specific discharge; drain discharge per unit area (mm/day)
ordinate of s-hydrograph (mm/day)
time of residence (s;day) return period (year)
time, period (s;day)
wetted perimeter of drain (m) 3
volume of reservoir (m )
flow velocity (m/day)
flow density (m/day)
soil moisture content at field capacity (mm)
entrance resistance (day/m)
radial resistance (day/m)
cartesian coordinates (m)
distance variable (m)
hydraulic head function at a distance x
from the drain; water depth in channel (m)
angle
partial derivative sign
small increment of
effective porosity; drainable pore space
dimensionless
LT"1
dimensionless
-1 LT
LT"1
T T
T
L
L3
LT"1
LT"1
L
L-'T
L-'T
L
L
L
degree
dimensionless
dimensionless
dimensionless
284
Literature
BERNSTEIN, L. 1974. Crop growth and salinity. Drainage for Agriculture,
Agronomy 17. Madison.
BOUSSINESQ, M.J. 1904. Recherches theoriques sur 1'ecoulement des nappes
d'eau infiltrees dans le sol et sur le debit des sources. Journal des
Mathematiques Pures et Appliquees, 10, 1-78.
BOUWER, H. 1974. Developing drainage design criteria. Drainage for
Agriculture. Agronomy 17. Madison.
BRAUN-BLANQUET and O.DE BOLOS. 1957. Les groupements vegetaux du Bassin
Moyen de l'Ebro et leur dynamisme. Anales de la Estacion Experimental
de Aula Dei. Zaragoza.
BURINGH, P. 1960. Soils and soil conditions in Iraq. Wageningen.
. 1970. Halomorphic soils. Introduction to the study of soils in
tropical and subtropical regions. PUDOC. Wageningen.
CAVELAARS, J.C. 1974. Subsurface field drainage systems. Drainage Principles
and Applications. Publ.16, Vol.IV. ILRI, Wageningen.
DE LOS RIOS, F. 1966. Colonizacion de las Bardenas, Cinco Villas, Somontano
y Monegros. Institucion Fernando El Catolico, Zaragoza.
DE MEESTER, T. 1970. Soils of the Great Konya Basin, Turkey. PUDOC,
Wageningen.
DE RIDDER, N.A. 1974. Groundwater survey. Drainage Principles and Appli
cations. Publ.16, Vol.III. ILRI, Wageningen.
DIELEMAN, P.J. (Ed.) 1963. Reclamation of salt affected soils in Iraq.
Soil hydrological and agricultural studies. ILRI, Publ.ll. Wageningen.
. 1974. Deriving soil hydrological constants from field drainage tests.
Drainage Principles and Applications. Publ.16, Vol.III. ILRI, Wageningen.
285
DRIESSEN, P.M. 1970. Soil salinity and alkalinity in the Great Konya Basin,
Turkey. PUDOC, Wageningen.
ELIAS CASTILLO, F., JIMENEZ ORTIZ, R. 1965. Evapotranspiraciones potencia-
les y balances de agua en Espana. Mapa Agronomico Nacional. Madrid.
FAO/UNESCO. 1973. Irrigation, Drainage and Salinity. An International
source book. London.
GOOSEN. 1967. Aerial photo-interpretation in soil survey. Soils Bulletin
No.6. FAO, Rome.
HERNANDEZ PACHECO, E. 1949. Las Bardenas Reales. Rasgos fisiograficos y
geologicos. Institucion Principe de Viana No.XXXVII. Pamplona.
IGME. Mapa Geologico de Espana 1:1 000 000. Sintesis geologica 1:200 000.
Hojas y Memorias del Mapa Geologico de Espana 1:50 000: Ejea, Sadaba,
Sos y Tauste. Instituto Geologico y Minero de Espana. Madrid.
ILRI. 1964. Code of practice for the design of open water courses and
ancillary structures. Bulletin 7. International Institute for Land
Reclamation and Improvement. Wageningen.
. 1972. Veldboek. Fieldbook for land and water management experts.
Wageningen.
IRYDA. 1973/1975. Reconocimiento detallado de los suelos de la Zona
Regable de las Bardenas. Zaragoza y Navarra. Seccion de Suelos del
Instituto Nacional de Reforma y Desarrollo Agrario. Madrid.
KESSLER, J. 1973. Field drainage criteria. Drainage Principles and
Applications. Publ.16, Vol.11. ILRI, Wageningen.
, DE RAAD, S.J. 1974. Analyzing rainfall data. Drainage Principles
and Applications. Publ.16, Vol.III. ILRI, Wageningen.
OOSTERBAAN, R.J., KESSLER, J. 1974. Determining hydraulic conductivity
of soils. Drainage Principles and Applications. Publ.16, Vol.III. ILRI,
Wageningen.
RH0ADES, J.D. 1974. Drainage for salinity control. Drainage for Agricul
ture. Agronomy 17. Madison.
286
RICHARDS, L.A. (Ed.) 1954. Diagnosis and improvement of saline and alkali
ne soils. Agr.Handbook 60. USDA, Washington D.C.
SAENZ GARCIA, C. 1942. Estructura general de la Cuenca del Ebro. Estudios
Geograficos. Ano III. Tomo 7. Madrid.
SIEBEN, W.H. 1974. Effect of drainage conditions on nitrogen supply and
yield of young loamy soils in the IJsselmeerpolders. Den Haag.
SOIL SURVEY STAFF. 1951. Soil Survey Manual. Agr.Handbook 18. USDA,
Washington D.C.
. 1975. Soil Taxonomy. Agr.Handbook 936. Washington D.C.
SOLE SABARIS. 1954. Sobre la estratigrafia de las Bardenas y los limites
del Oligoceno y Mioceno en el Sector Occidental de la depresion del
Ebro. Real Sociedad Esp. de Historia Natural. Madrid.
TALSMA, T. 1963. The control of saline groundwater. Mededelingen van
de Landbouwhogeschool. Wageningen.
VAN DER MEER, K., MESSEMAECKERS VAN DE GRAAFF. 1974. Hydropedological
Survey. Drainage Principles and Applications. Publ.16, Vol.III. ILRI,
Wageningen.
VAN DER MOLEN, W.H. 1972. Water management (drainage). Syllabus M.Sc.
Course on Soil Science and Water Management. Agric.University, Wageningen.
— • — . 1972. Non-steady groundwater flow. Syllabus M.Sc.Course on Soil
Science and Water Management. Agric.University, Wageningen.
and J.W.van HOORN. 1976. Salt balance and leaching requirement. Syl
labus of the 15th International Course on Land Drainage. ILRI,Wageningen.
VAN HOORN, J.W. 1960. Groundwater flow in basin clay soil and the deter
mination of some hydrological factors in relation with the drainage
system. PUDOC, Wageningen.
. 1974. Drainage of heavy clay soils. Drainage Principles and Appli
cations. Publ.16, Vol.IV. ILRI, Wageningen.
VEENENBOS, J.S. 1972. Soil Maps and Soil Survey Interpretation. Syllabus
of the M.Sc.Course on Soil Science and Water Management. Agric.Uni
versity, Wageningen.
287
WESSELING, J. 1973. Subsurface flow into drains. Drainage Principles and
Applications. Publ.16, Vol.11. ILRI, Wageningen.
WITHERS, B., and VIPONS, S. 1954. Irrigation design and practice.
Batsford. London.
288
Summary
C h a p t e r 1
The Ebro basin is situated in north-eastern Spain and forms a geo
graphic unit bounded by high mountains. The Bardenas area lies in the Ebro
basin and forms part of the Bardenas Alto - Aragon irrigation scheme, which
was designed to make use of the surface water resources from the Pyrenees.
Chapter 2
The Ebro basin is a tertiary sedimentation basin in which the Ebro
river and its main tributaries have incised alluvial valleys. The tertiary
sediments consist mainly of mudstone, locally with interbedded gypsum
layers, and very fine siltstone. Both sedimentary rocks are fine textured
and, because they were deposited in a brackish lacustrine environment,
contain harmful soluble salts.
The main landscape-forming processes were erosion, transport, and
deposition under semi-arid climatic conditions. The highest parts of the
landscape consist of old tertiary formations which form the uplands of a
dissected plain. At a lower level mesas occur, which consist of coarse
alluvium covering the underlying tertiary sediments. Most of the eroded
sediments were removed from the area but local sedimentation also occurred.
Owing to the semi-arid conditions, both sediments and salts were deposited.
The highest salt concentrations are found in the lowest parts of the allu
vial formations, especially where the alluvium was derived from the eroded
mudstone and siltstone. Between the residual uplands and the low-lying
alluvial formations, piedmont and colluvial slopes occur.
289
Within the Bardenas area ten major physiographic units were defined,
each of them subdivided into minor components and indicated on the soil
map.
The Ebro basin is the driest part of northern Spain. The climate is
semi-arid and becomes drier from the borders to the centre of the depression.
The seasonal variation in temperature is great. Potential evapotranspi-
ration exceeds total precipitation, which is extremely variable and is not
concentrated in distinct rainy seasons. Wind velocity is high and both cold
and warm dry winds are common. Evaporation thus occurs even in winter when
temperatures are low.
A great part of the area is cultivated, so that natural vegetation
is restricted to residual and eroded soils not used for agriculture and to
salt-affected soils where halophytes grow.
Irrigated farming is influenced by soil conditions. Salt-free soils
are under full irrigation, the main crops being maize, lucerne, sugar beet,
and some horticultural crops.
The cropping pattern on the saline soils depends on the degree of sal
inity. Barley and sugar beet are grown on moderately saline soils and lucerne
on succesfully leached soils. On the higher lands, not under the command of
the irrigation scheme, barley is grown.
Chapter 3
The study area comprises two drainage basins. The northern part drains
to the Aragon river, the southern part to the Riguel river, which is a trib
utary of the Arba river.
Drainage and salinity of the groundwater depend on the situation of
each geomorphological unit and its relation to adjacent units. The ground
water in the fluvio-colluvial formations of the northern basin is shallow
and highly saline. An ephemeral perched water table is found in the mesas,
290
where the groundwater is non-saline. No shallow water table was found in
other physiographic units.
The irrigation water is of good quality as its EC is at the lower end
le C„-range. The SAR is also in the
so there is no danger of alkalinization.
of the C„-range. The SAR is also in the lowest range S and the RSC is zero,
Chapter 4
The physiographic approach was used to prepare the soil map. Each
mapping unit is a broad association of soils having similar salinity hazards
and possibilities of reclamation.
Five main soil associations were distinguished:
a) The residual soils of the siltstone outcrops, which have only
a thin surface horizon overlying the hard siltstone.
b) The soils of the mesas, which consist of a reddish loamy surface
horizon overlying semi-consolidated coarse alluvium rich in calcium-carbo
nate but free of other salts. This in turn overlies the impervious mudstone.
Texture and depth of the soil profile vary. Where moderately deep soils
occur, a prosperous irrigated agriculture flourishes.
c) The soils of the piedmont and colluvial slopes were developed
from a mixture of fine colluvium and material from the underlying tertiary
sediments. They are generally deep and fine textured and have an intrinsic,
though variable, salinity, increasing with depth. Because of the low permeab
ility and the salinity of the subsoil, the most suitable irrigation method
is sprinkling.
d) The non-saline soils of the alluvial valleys of the main rivers.
Soil conditions vary greatly, but the older terrace soils are usually shal
lower and less suitable for irrigation than the youngest deeper (alluvial)
soils. In general, prosperous irrigated agriculture exists on these soils.
e) The saline alluvial and fluvio-colluvial soils of valleys and fans,
whose parent material was derived from denudation of the tertiary sediments.
Soil conditions and the degree of salinity vary in each mapping unit, and
consequently the possibilities of reclamation vary as well.
291
Piezometers were installed to monitor the water table. Precipitation
was measured, as were the amounts of irrigation and drainage water. Soil
samples were taken at fixed sites to determine the salinity during the
leaching process.
Chapter 8
At the Valarena field, water flowed directly into the drain trench
through the upper layer of soil, in which the stratification had been
disrupted by levelling and subsoiling. Below this layer, there was no per
colation of water and therefore no desalinization. These soils cannot be
leached merely by the provision of a drainage system but also require deep
subsoiling.
At the Alera field, unsteady groundwater flow prevailed. At the end of
tail recession, flow conditions approached those of steady flow. The dis
charge/hydraulic head relation had a parabolic shape showing that flow was
restricted to the soil above drain level because the drains had been placed
just above the impervious layer.
The Boussinesq theory was very suitable to study the drainage of the
Alera field. At the end of tail recession, if the term for flow below drain
level was disregarded, the Hooghoudt equation could be applied.
Drainable pore space was determined from the fall of the water table
and the amount of drainage water during periods of low evapotranspiration.
An average value of 4 per cent was found.
The hydraulic conductivity was calculated from the discharge/hydraulic
head relation using the Boussinesq and Hooghoudt equations for periods of
low evapotranspiration. In general good agreement was found among the values
obtained. It could thus be concluded that:
Hydraulic conductivity decreases with depth, becoming negligible
below drain level.
The hydraulic conductivity between a depth of 0.5 m and drain level
equals about 0.6 m/day, and is about 1.5 m/day in the upper layer.
294
- For high water table conditions, the average hydraulic conductivity
of the soil profile is 1 m/day.
A comparison of hydraulic conductivity values obtained with field and
laboratory methods and those obtained from the discharge/hydraulic head re
lation showed that:
The results obtained with the auger hole method (K - 0.2 m/day) were
lower than those derived from the discharge/hydraulic head relation.
No satisfactory results were obtained from the inversed auger hole
measurements above the water table.
The results obtained from laboratory measurements in undisturbed
soil cores showed the anisotropy of the soil.
The entrance resistance (W ) of different combinations of drainage
and filter materials was calculated from the drain discharge and the head
loss of the water table measured in the drain trench (h.). Another method,
by which the head loss in the trench was calculated from the shape of the
water table was also applied. Both methods gave similar results, yielding
the following conclusions:
- The W -values remained fairly constant with time, except for plastic
pipes with an envelope of esparto or coconut fibre for which an increase in
W was observed, e
- The best combination was clay pipes with a gravel cover (W =2 day/m).
Corrugated PVC-pipes with gravel covering and clay pipes without
gravel may be used also (W = 5 day/m).
Corrugated plastic pipes without a filter gave less satisfactory
results (W = 1 3 day/m).
Plastic pipes with coconut fibre and esparto filters showed an
even higher W than plastic pipes without a filter.
Barley straw is an unsuitable cover material since it rots easily
and clogs the pipe.
295
Chapter 9
The desalinization of the Alera field started with an initial leaching,
followed by the irrigated cultivation of moderately salt-resistant crops.
The leaching efficiency coefficient was determined by comparing the
actual desalinization process with theoretical models. Thus the leaching
requirement could be predicted for different initial salt contents.
To exclude the influence of slightly soluble salts, the desalinization
curves were drawn in terms of chloride content. The correlation between
chloride percentage and electrical conductivity was high.
The following conclusions emerged from the study of the leaching
process:
- The leaching efficiency coefficient was not constant but increased
with depth.
- The leaching efficiency coefficient was higher at the beginning of
the desalinization process and decreased gradually as the soil became less
saline.
- The calculated values reflected the differences in soil structure.
An average value of 0.5 was determined for the upper layer (0-50 cm),
and a value of 1.0 for the deeper layer (50-100 cm).
- The initial salinity was related to soil physical properties
(infiltration rate and permeability) which, in turn, were dependent on the
compactness of the soil.
For an initial EC of 15 mmhos/cm, approximately 1000 mm of perco
lation water are required, which meant 1100 to 1400 mm of irrigation water.
The leaching period could last up to 8 months, from early autumn to late May.
- Deep subsoiling and local gypsum applications improved the structure
of the upper soil layer.
The leaching of saline soils could be split into two phases: an
initial leaching of the upper layer, followed by the irrigated cultivation
296
of a moderately salt-resistant crop (sugar beet), during which percolation
losses leached the deeper layers.
- There was no risk of alkalinization during the leaching period.
- To prevent secondary salinization after reclamation, good drainage
conditions must be maintained and an irrigated crop rotation practised.
Chapter 10
From the relation between the depth of the water table, crop growth,
and the mobility of agricultural machinery on the soil, and from a study
of the groundwater regime in winter and during the irrigation season drainage
criteria for unsteady-state conditions were derived. These criteria were
converted to steady-state criteria for easier use in drainage projects.
The following conclusions could be drawn from the study:
Little harm is done to winter cereals if a water table remains
within a depth of 50 cm for no more than 3 consecutive days.
- With a water table between 75 and 100 cm, sugar beet grows well
and is not harmed if a water table is within the top 50 cm for 3 to A con
secutive days.
- Lucerne is more sensitive than sugar beet to high water tables. For
good yields, a water table must not remain longer than 3 days within the
top 25 cm of soil, 4 or 5 days within the top 50 cm, and 5 or 6 days within
the top 75 cm.
A water table depth shallower than 65 cm prevents the movement of
machinery and hampers seed-bed preparation in winter.
The following drainage criteria were assessed:
a) In winter, a water table drawdown from the soil surface to a depth
of 0.65 m in 8 days.
b) In the irrigation season, a water table rise of 0.7 m caused by
irrigation losses must be lowered in the 12 days between two consecutive
irrigations, and must be deeper than 0.7 m after 7 days.
297
Applying these criteria in the Boussinesq equation for unsteady flow
and using the values for hydraulic conductivity and drainable pore space
determined at the experimental field, a spacing of 25 m for drains installed
at a depth of 1.2 m was obtained.
Equivalent drainage criteria for steady flow are a minimum depth of
0.5 m for the unsaturated zone, with a corresponding hydraulic head midway
between drains of 0.7 m and a drain discharge of 3 mm/day.
If the entrance resistance was taken into account, the Ernst equation
for steady flow and the Hellinga/de Zeeuw equation for unsteady flow could
be used in calculating the drain spacing. The results obtained by both
approaches agree well and allowed the following conclusions:
For drainage and filter materials with a high entrance resistance,
the drain density (m/ha) required becomes twice that needed for materials
with low entrance resistance.
Material with high entrance resistance involves much more risk of
failure than a wider spacing with good material.
298
Resumen
C a p i t u l o 1
La cuenca del Ebro esta situada en el nordeste de Espana y forma una
unidad geografica rodeada por altas montanas. La zona de las Bardenas esta
situada en la cuenca del Ebro y forma parte del plan de riegos Bardenas-Alto
Aragon, proyectado para el aprovechamiento de los recursos hidraulicos pro-
cedentes de los Pirineos.
Capitulo 2
La depresion del Ebro es una cuenca de sedimentacion terciaria en la
que el rio Ebro y sus afluentes principales han formado valles aluviales.
Los sedimentos terciarios consisten en margas limosas, a veces con capas de
yeso intercaladas y areniscas de grano muy fino. Las dos rocas sedimenta-
rias son de textura fina y contienen sales solubles debido a que fueron de-
positadas en un medio lacustre salobre.
El principal proceso remodelador del paisaje ha sido la erosion segui-
do de transporte y deposicion bajo condiciones climaticas semiaridas. Las
formaciones terciarias mas antiguas ocupan las posiciones mas altas del
paisaje, formando las tierras altas de una llanura diseccionada. A un nivel
mas bajo existen mesas formadas por un aluvio grueso que descansa sobre se
dimentos terciarios subyacentes. La mayor parte de los sedimentos erosionados
fueron transportados fuera del area, pero tambien ha habido sedimentacion
local. Debido a las condiciones semiaridas, tanto los sedimentos como las
sales transportadas fueron depositadas juntas. Por tanto las mayores concen-
traciones de sales se encuentran en las partes mas bajas de las formaciones
aluviales, especialmente en las que el aluvio procede de las margas y arenis
cas erosionadas. Entre las tierras altas residuales y las formaciones alu
viales mas bajas existen laderas de piedemonte y laderas coluviales.
299
e) Los suelos salinos aluviales y fluvio-coluviales de los valles y
abanicos, en los que el material parental procede de la denudacion de los
sedimentos terciarios. Las condiciones de los suelos y el grado de salinidad
varian en cada unidad cartografica y consecuentemente las posibilidades de
recuperacion varian tambien.
Capxtulo 5
El origen de las sales es por tanto la salinidad intrinseca de estos
materiales parentales y la salinizacion secundaria en zonas que reciben
agua y que carecen de drenaje natural. Bajo riego la movilizacion y redis
tribution de sales continua y ciertamente aumenta.
Los suelos salinos de la zona estan afectados principalmente por clo-
ruro sodico, componente dominante en todas las muestras. Ademas en la
cuenca norte existen sulfato de calcio y magnesio, mientras que en la sur,
los cloruros de calcio y magnesio predominan sobre los sulfatos.
El SAR aumenta a medida que la EC lo hace. La alcalinidad del suelo
puede considerarse como un reflejo de la salinidad, ya que los suelos muy
salinos son sodicos tambien. Suelos alcalinos no-salinos no han sido en-
contrados y valores de pH mayores de 8.5 no existen.
Los resultados sobre la tolerancia de los cultivos a la salinidad,
obtenidos en pruebas de campo corresponden bien con los niveles de tole
rancia generalmente aceptados.
La utilizacion permanente de los suelos ligeramente salinos puede
asegurarse manteniendo el sistema de drenaje actual de zanjas abiertas y
drenes interceptores, y manteniendo los suelos bajo riego total. Las perdi-
das por percolacion inherentes al riego de gravedad son suficientes para
lavar las sales depositadas en la zona radicular.
El riego por aspersion es una solucion apropiada para los suelos de
las laderas, ya que no requiere nivelacion y las pequenas cantidades de
agua de riego que se aplican reducen las filtraciones de agua salina. El
sistema de drenaje requerido se reduce a un dren interceptor entre la la-
dera y el valle adyacente.
302
Capltulo 6
Los suelos aluviales salinos requieren recuperacion. Para ello debe
instalarse un sistema de drenaje, y a continuacion practicar un lavado ini-
cial para reducir el contenido de sales.
Como no habia experiencia local sobre los procesos de drenaje y desa-
linizacion, se llevo a cabo un proceso experimental de recuperacion. En
consecuencia se seleccionaron dos fincas experimentales.
La finca de Alera representa los suelos mal drenados de las formaciones
fluvio-coluviales de la cuenca norte. Son suelos arcillo-limosos cuya poro-
sidad y permeabilidad decrece con la profundidad. Por debajo de una profun-
didad media de 1,5 m el suelo llega a ser casi impermeable. La salinidad
aumenta con la profundidad, llegando a ser de 20 a 35 mmhos/cm en la capa
menos permeable. En la capa superficial la variacion es mayor.
La finca de Valareiia representa los suelos salinos de los valles y aba-
nicos aluviales de la cuenca de drenaje sur. Son suelos franco-arcillo-limo-
sos con una marcada estratificacion muy fina. A una profundidad media de 2,5
m existe un aluvio grueso saturado de agua muy salina,que descansa sobre una
marga impermeable.Debido a la estratificacion la conductividad hidraulica es
anisotropica. La salinidad del suelo esta distribuida mas uniformemente que
en los suelos de Alera.
El proceso de recuperacion consistio en las fases siguientes:
a) Diseno teorico del sistema de drenaje, basado en las propiedades hidrologicas del suelo medidas por metodos de campo conveneionales y en criterios de drenaje supuestos.
b) Realizacion del sistema diseiiado en las parcelas experimentales.
c) Recopilacion de los datos de campo y determinacion de las propiedades hidrologicas del suelo y de los criterios de drenaje reales.
d) Diseno del sistema de drenaje definitivo que pueda ser la base de las recomendaciones para la recuperacion de suelos salinos con condiciones similares.
303
lares, deduciendose las siguientes conclusiones:
- Los valores de W permanecieron constantes, excepto en la tuberia
de plastico con envolvente de fibra de coco y esparto en los que se
observo un incremento de W . e
- La mejor combinacion consiste en tuberia de ceramica con filtro de
grava (W = 2 dias/m).
Tambien pueden utilizarse tuberias de PVC corrugado con filtro de
grava y tuberias de ceramica sin grava (W = 5 dias/m).
- La tuberia de PVC corrugado sin filtro ha mostrado resultados menos
satisfactorios (W = 1 3 dias/m). e
- Las tuberias de plastico con filtros de esparto y coco han mostrado
una resistencia aun mas alta que las tuberias sin filtro.
- La paja de cebada no es adecuada como material envolvente ya que
se descompone facilmente y obtura la tuberia.
Capitulo 9
El proceso de desalinizacion de los suelos de la finca de Alera comenzo
con un lavado inicial seguido de riego de cultivos resistentes a la salini-
dad.
El coeficiente de eficacia de lavado fue determinado por comparacion
del proceso de desalinizacion real con modelos teoricos. De este modo pueden
hacerse predicciones de las necesidades de lavado para diferentes contenidos
salinos iniciales.
Para evitar la influencia de sales ligeramente solubles las curvas de
desalinizacion fueron dibujadas en terminos de contenido de cloruros. La
correlacion entre el porcentaje de cloruros y la conductividad electrica
es alta.
306
Del estudio del proceso de lavado se dedujeron las siguientes conclu-
siones:
- El coeficiente de lavado no fue constante, sino que aumentaba con
la profundidad.
- El coeficiente de lavado fue mayor al comienzo del proceso de desa-
linizacion y decrecio gradualmente a medida que el suelo era menos
salino.
- Los valores calculados reflejaron las diferencias en estructura del
suelo.
- En la capa superficial (0 - 50 cm) se determino un valor medio de
0.5, y para la capa muy profunda (50 - 100 cm) un valor de 1.0.
- El nivel de salinidad inicial estaba relacionado a propiedades fi-
sicas del suelo (velocidad de infiltracion y permeabilidad), que
dependen a su vez de la compacidad del suelo.
- Para una EC inicial de 15 mmhos/cm se necesitan unos 1.000 mm de e
agua de percolacion, lo que significa de 1.100 a 1.400 mm de agua
de riego. El periodo de lavado puede durar unos 8 meses, desde el
comienzo del otono hasta finales de Mayo.
- Un subsolado profundo en union de aplicaciones locales de yeso,
mejoraron la estructura de la capa superior del suelo.
- El lavado de suelos altamente salinos podra dividirse en dos fases:
un lavado inicial de la capa superficial, seguido de cultivo de
plantas moderadamente resistentes a la salinidad (remolacha azuca-
rera), durante el cual las perdidas por percolacion continiian el
lavado de las capas mas profundas.
- No hubo riesgo de alcalinizacion durante el periodo de lavado.
Para impedir la salinizacion secundaria despues de la recuperacion,
deben mantenerse condiciones de drenaje libre y debe practicarse una
rotacion de cultivos bajo riego.
307
Capltulo 10
Los criterios de drenaje para regimen variable se han deducido del
estudio de la relacion entre la profundidad de la capa de agua, el crecimien-
to de los cultivos y la movilidad de maquinaria agricola sobre el suelo y
del estudio del regimen del agua freatica en invierno y durante la estacion
de riego. Estos criterios fueron reducidos a criterios de regimen permanen-
te, para una utilizacion mas sencilla en proyectos de drenaje. Del estudio
pueden deducirse las conclusiones siguientes:
- Los cereales de invierno no presentan danos considerables si la
capa freatica permanece a menos de 50 cm de la superficie durante
3 dias consecutivos.
- La remolacha azucarera se desarrolla bien con una capa de agua per-
manente entre 75 y 100 cm, siempre que la capa de agua no permanezca
en los 50 cm mas superficiales durante mas de 3 6 4 dias consecuti
vos.
- La alfalfa es con mucho mas sensible a capas freaticas altas. Se
obtuvo una buena produccion cuando la capa freatica permanecio
menos de 3 dias en los 25 cm superficiales, durante 4 5 5 dias
por encima de 50 cm y de 5 a 6 dias por encima de 75 cm.
- La existencia de una capa de agua por encima de 65 cm impide el
movimiento de maquinaria y por tanto la preparaci6n de la siembra
en invierno.
Los criterios de drenaje fueron definidos de la siguiente forma:
a) En invierno, debe conseguirse un descenso de la capa de agua desde la misma superficie hasta una profundidad de 0.65 m en 8 dias.
b) En la estacion de riego debe controlarse una elevacion de la capa de agua de 0.7 m, debida a perdidas de riego, en un periodo de 12 dias entre dos riegos consecutivos, y debe estar por debajo de una profundidad de 0.7 m despues de 7 dias.
308
Aplicando estos criterios a la ecuacion de Boussinesq para flujo vari
able y utilizando los valores de la conductividad hidraulica y del espacio
poroso drenable determinados en la finca experimental, se obtuvo un espa-
ciamiento de 25 m para drenes instalados a una profundidad de 1.2 m.
Los criterios de drenaje equivalentes para regimen permanente son:
una profundidad minima de la zona radicular no saturada de 0.5 m, con una
correspondiente carga hidraulica en el punto medio entre dos drenes de 0.7 m
y una descarga de 3 mm/dia.
Si se tuviera en cuenta la resistencia de entrada podrian utilizarse
la ecuacion de Ernst para regimen permanente y la ecuacion de Hellinga-
de Zeeuw para regimen variable para el calculo del espaciamiento entre dre
nes. Los resultados obtenidos con ambas aproximaciones mostraron una buena
concordancia y permitieron deducir las siguientes conclusiones:
Con materiales de drenaje y filtros que presentan altas resistencias
de entrada, la densidad de drenes (m/ha) requerida llega a ser doble que la
requerida con drenes de baja resistencia de entrada.
- Un material con alta resistencia de entrada lleva consigo mucho mas
riesgo de fracaso, que un mayor espaciamiento con un buen material.
309
Samenvatting
Hoofdstuk 1
Het stroomgebied van de Ebro ligt in het noordoosten van Spanje en
vormt een geografische eenheid die door hoge bergen wordt begrensd. Het
Bardenas gebied ligt binnen het stroomgebied van de Ebro en maakt deel uit
van het Bardenas-Alto Aragon irrigatieproject dat zijn oppervlaktewater uit
de Pyreneeen betrekt.
Hoofdstuk 2
Het Ebro stroomgebied is een tertiair sedimentatiebekken waarin de Ebro
en zijn zijrivieren alluviale dalen hebben gevormd. De tertiaire afzettin-
gen bestaan hoofdzakelijk uit moddersteen, waarin plaatselijk lagen van gips
en zeer fijne siltsteen zijn ingeschakeld. Beide afzettingsgesteenten bezit-
ten een fijnkorrelige textuur en bevatten schadelijke oplosbare zouten,omdat
zij werden afgezet in een lacustrisch brakwater milieu.
Het belangrijkste landschapsvormende processen waren erosie, transport
en afzetting onder semi-aride klimaatsomstandigheden. De hoogste delen van
het landschap bestaan uit oud-tertiaire formaties, die plateaus vormen in
een door rivieren doorsneden vlakte. Op een lager niveau komen mesas voor,
die bestaan uit grof alluviaal materiaal dat de onderliggende tertiaire af-
zettingen bedekt. Het meeste geerodeerde materiaal werd uit het gebied af-
gevoerd, maar plaatselijk vond ook sedimentatie plaats. Tengevolge van de
semi-aride omstandigheden kwamen eerder genoemde sedimenten en zouten tot
afzetting. De hoogste zoutgehalten worden dan ook aangetroffen in de laagst-
gelegen delen van de alluviale afzettingen, in het bijzonder daar waar het
alluviale materiaal afkomstig is van de geerodeerde moddersteen en siltsteen.
Tussen de voor erosie gespaard gebleven plateaus en de laaggelegen alluviale
311
afzettingen komen piedmont-vlakten en colluviale hellingen voor.
In het Bardenasgebied werden tien grote fysiografische eenheden onder-
scheiden, die in kleinere eenheden werden onderverdeeld en op een bodemkaart
aangegeven.
Het Ebrobekken is het droogste gebied van Noord-Spanje. Het klimaat
is semi-aride en wordt naar het centrum van het bekken droger. De tempera-
tuur vertoont een grote seizoensschommeling. De potentiele verdamping over-
treft de neerslag die van jaar tot jaar grote verschillen te zien geeft en
niet geconcentreerd is in duidelijke regenseizoenen. De windsnelheid is hoog
en zowel warme als koude droge winden komen voor. De verdamping is daarom
ook in de winter, wanneer de temperatuur laag is, niet te verwaarlozen.
In een groot gedeelte van het gebied wordt landbouw uitgeoefend, zodat
de natuurlijke begroeiing beperkt is tot de geerodeerde plateaus die niet
voor de landbouw worden gebruikt en tot de zoute gronden die een vegetatie
van halofyten bezitten.
Bodemomstandigheden zijn bepalend voor de vorm van geirrigeerde land
bouw. Volledige irrigatie wordt toegepast op niet-zoute gronden, met mais,
luzerne, suikerbieten en enige tuinbouwgewassen als de voornaamste gewassen.
Het gewassenpatroon op de zoute gronden hangt af van het zoutgehalte van
de grond. Gerst en suikerbieten worden op matig zoute gronden verbouwd,
luzerne op gronden die met goed resultaat zijn ontzout. Op de hoger gelegen
gronden, die buiten het irrigatiegebied vallen, wordt gerst verbouwd.
Hoofdstuk 3
Het bestudeerde gebied watert naar twee zijden af, het noordelijke
gedeelte naar de Aragon en het zuidelijke naar de Riguel, een zijrivier
van de Arba.
De afwatering en het zoutgehalte van het grondwater zijn afhankelijk
van de ligging van de verschillende geomorfologische eenheden ten opzichte
van aangrenzende eenheden. In de noordelijk gelegen fluvio-colluviale af
zettingen komt ondiep, zeer zout grondwater voor. Een tijdelijke grondwater-
312
spiegel op geringe diepte wordt aangetroffen in de mesas, waar het grond-
water zoet is. In de andere fysiografische eenheden komen geen ondiepe
grondwaterspiegels voor.
Het irrigatiewater is van goede kwaliteit: de EC-waarde bedraagt onge-
veer 0,3 mmhos/cm. Er bestaat geen gevaar voor alkalinisatie, want de
SAR-waarde varieert van 0,2 tot 1,3 en de RSC-waarde is nul.
Hoofdstuk 4
Volgens de fysiografische benaderingswijze werd een bodemkaart van
het onderzoeksgebied samengesteld. Iedere kaarteenheid bestaat uit een
groepering van bodemsoorten met overeenkomstige verzoutingsgevaren en mo-
gelijkheden tot ontginning. De volgende vijf hoofdbodemgroepen werden
onderscheiden:
a) De residuaire gronden van de dagzomende siltsteenafzettingen, die
gekenmerkt worden door een dunne oppervlaktelaag gelegen op harde siltsteen.
b) De mesa-gronden die bestaan uit een roodachtige, lemige bovenlaag
gelegen op ten dele verkit, grof alluviaal materiaal dat rijk is aan calcium-
carbonaat, maar waarin andere zouten ontbreken. Aan de onderzijde wordt dit
materiaal begrensd door ondoorlatende moddersteen. De textuur en de diepte
van het bodemprofiel varieren. Op de matig diepe gronden komt een welvarende,
ge'irrigeerde landbouw voor.
c) De gronden van de piedmont-vlakten en colluviale hellingen ont-
wikkelden zich uit een mengsel van fijn colluvium en materiaal van de onder-
liggende tertiaire afzettingen. Deze, doorgaans diepe, gronden hebben een
fijne textuur en bevatten zout waarvan de concentratie van plaats tot plaats
wisselt maar met de diepte toeneemt. In verband met de geringe doorlatend-
heid en het zoutgehalte van de ondergrond, is beregening de meest geschikte
irrigatiemethode.
d) De niet-zoute gronden van de alluviale dalen der hoofdrivieren.
De bodemomstandigheden wisselen sterk, maar gewoonlijk zijn de oudere terras-
gronden ondieper en minder geschikt voor irrigatie dan de jongere, diepere,
313
alluviale gronden. In het algemeen is op deze gronden welvarende geirri-
geerde landbouw.
e) De zoute, alluviale en fluvio-colluviale gronden van dalen en puin-
kegels waarvan het moedermateriaal afkomstig is van geerodeerde tertiaire
afzettingen. De bodemomstandigheden en het zoutgehalte verschillen binnen
elke kaarteenheid en derhalve ook de mogelijkheden tot ontginning.
Hoofdstuk 5
De herkomst van het zout moet gezocht worden in het zoute moederge-
steente. Secundaire verzouting treedt op in gebieden die water ontvangen,
maar geen natuurlijke drainage bezitten. Wanneer in deze gebieden wordt
geirrigeerd, gaan oplossing, transport en herverdeling van zouten door en
neemt de verzouting toe.
Natriumchloride is het meest voorkomende zout in de bodem. Bovendien
zijn in het noordelijke gedeelte van het gebied calcium- en magnesiumsulfaat
aanwezig, terwijl in het zuidelijke deel meer calcium- en magnesiumchloride
voorkomen.
De SAR-waarde neemt toe met de EC-waarde. Alkaliniteit kan derhalve
beschouwd worden het zoutgehalte van de grond te weerspiegelen daar zeer
zoute gronden ook een hoog gehalte aan natrium bezitten. Niet-zoute alkali-
gronden komen niet voor, evenmin als pH-waarden die hoger zijn dan 8,5.
De resultaten van veldproeven betreffende de zouttolerantie van gewassen
komen goed overeen met de algemeen aanvaarde normen voor zout-tolerantie.
Het blijvend in gebruik houden van de zwak-zoute gronden (EC = 4 tot
8 mmhos/cm) kan verzekerd worden door het huidige drainage stelsel van open
sloten en interceptor drains te handhaven en de gronden regelmatig te irri-
geren. De normale percolatieverliezen die bij basisirrigatie optreden, zijn
voldoende om de zouten uit de wortelzone te spoelen.
Beregening is geschikt voor de gronden op de hellingen, omdat in dat
geval geen egalisatie nodig is en de kleine watergiften de stroming van zout
water naar aangrenzende gebieden beperken. Slechts een interceptor drain
tussen de helling en het dal is nodig.
314
Hoofdstuk 6
Voor de ontginning van de zoute gronden moet een drainagestelsel wor-
den aangelegd waarna doorspoeling moet plaats vinden om het zoutgehalte te
verlagen. Plaatselijke ervaring met drainage en ontzilting ontbrak, reden
waarom werd besloten tot het nemen van proeven. Voor dit doel werden twee
proefvelden uitgekozen.
Het Alera proefveld, gelegen op de fluvio-colluviale afzettingen van
het noordelijke gedeelte van het gebied, werd representatief geacht voor de
slecht ontwaterde gronden in dit gebied. Deze gronden bestaan uit zware klei
waarvan het pori'envolume en de doorlatendheid met de diepte afnemen. Beneden
een diepte van 1,5 m is de grond vrijwel ondoorlatend. Het zoutgehalte neemt
met de diepte toe en bereikt in de vrijwel ondoorlatende laag waarden over-
eenkomende met 20 tot 35 mmhos/cm in elektrische geleidbaarheid. Het zout
gehalte in de bovenlaag varieert sterk.
Het Valarena proefveld ligt op de zoute gronden van de alluviale dalen
en puinkegels in het zuidelijke deel van het gebied.Deze gronden bestaan uit
lichte klei met een uitgesproken laagsgewijze opbouw.Op een diepte van 2,5 m
ligt de overgang naar grofkorrelig alluviaal materiaal, dat met zeer zout
grondwater verzadigd is en dat rust op de ondoorlatende moddersteen. Als
gevolg van de sterke gelaagdheid vertoont de doorlatendheid een duidelijke
anisotropie. De bodemverzouting komt veel regelmatiger verdeeld voor dan in
het Alera proefgebied.
Het onderzoek bestond uit de volgende stappen:
a) Theoretisch ontwerp van het drainagestelsel, waarbij gebruik ge-
maakt werd van de hydrologische bodemeigenschappen die werden bepaald vol-
gens conventionele veldmethoden, en van aangenomen drainage criteria.
b) Aanleg van de drainagestelsels in de beide proefvelden.
c) Verzamelen van veldgegevens gevolgd door bepaling van de werkelijke
hydrologische bodemeigenschappen en drainage criteria uit deze gegevens.
d) Ontwerp van het definitieve drainagestelsel dat als uitgangspunt
zal dienen voor het doen van aanbevelingen voor de ontginning van zoute
gronden onder soortgelijke omstandigheden.
315
Hoofdstuk 7
Na een gedetailleerde hydrologische en bodemkundige kartering werd voor
beide proefvelden een drainafstand van 20 m bij een draindiepte van 1,5m
berekend. Om de geringe infiltratiesnelheid te verbeteren, werden beide
velden gewoeld tot een diepte van 0,5 m. Piezometers werden geplaatst om
de veranderingen in de grondwaterstand te kunnen waarnemen. De hoeveelheden
irrigatie- en drainagewater werden gemeten, evenals de neerslag. Op vaste
plekken werden bodemmonsters genomen ter bepaling van het zoutgehalte
tijdens de doorspoeling.
Hoofdstuk 8
Op het Valareiia proefveld stroomde het water direct door de bovenlaag
naar de drainsleuf, omdat de gelaagdheid door woelen en egaliseren van
deze laag verbroken was. Beneden deze laag vond geen percolatie van water
plaats en derhalve geen ontzilting. Voor het doorspoelen van deze gronden
is niet alleen een drainagestelsel nodig, maar zij moeten ook tot grotere
diepte gewoeld worden.
Op het Alera proefveld verliep de grondwaterstroming niet-stationair.
Tegen het einde van het staartverloop werd een stationaire stroming benaderd.
Uit het verband tussen afvoer en potentiaal, dat een parabolische vorm had,
bleek dat de stroming beperkt bleef tot het profiel boven drainniveau en
dat de drains op een slecht doorlatende laag gelegen waren.
De Boussinesq theorie bleek zeer geschikt voor de bestudering van de
drainage in het Alera proefgebied. Aan het eind van het staartverloop kon de
formule van Hooghoudt worden toegepast, onder weglating van de term voor de
stroming beneden drainniveau.
De bergingscoefficient werd bepaald uit de daling van de grondwater
stand en de hoeveelheid drainagewater gedurende perioden met lage verdamping.
De gemiddelde waarde bedroeg 4%.
De doorlatendheid werd berekend uit het verband tussen afvoer en poten
tiaal met behulp van de formules van Boussinesq en Hooghoudt voor perioden
316
met lage verdamping. In het algemeen bleken de berekende waarden onderling
goed overeen te stemmen. Uit de resultaten van dit onderzoek konden de
volgende conclusies worden getrokken:
- De doorlatendheid neemt af met de diepte en is verwaarloosbaar klein
beneden drainniveau.
- De doorlatendheid bedraagt 0,6 m/dag tussen een diepte van 0,5 m
en drainniveau en ongeveer 1,5 m/dag in de bovenlaag.
- De gemiddelde doorlatendheid is 1 m/dag wanneer hoge grondwater-
standen voorkomen.
Uit een vergelijking van doorlatendheidswaarden verkregen met veld- en
laboratoriummethoden enerzijds en uit het verband tussen afvoer en poten-
tiaal anderzijds is gebleken dat:
de waarden gemeten met de boorgatenmethode (0,2 m/dag) lager waren
dan die berekend uit het verband tussen afvoer en potentiaal;
geen bevredigende resultaten werden verkregen met de omgekeerde
boorgatenmethode toegepast op de onverzadigde zone;
de metingen aan ongeroerde grondmonsters in het laboratorium op een
anisotropic van de grond wezen.
De intreeweerstand (W ) van verschillende combinaties van drain- en fil-e
termaterialen werd berekend uit de drainafvoer en het potentiaal-verschil
(h.) gemeten tussen de rand van de drainsleuf en de drainbuis. Een andere
methode, waarbij dit potentiaalverschil berekend wordt uit de vorm van de
grondwaterspiegel, werd eveneens toegepast. Beide methoden leverden vrijwel
dezelfde uitkomsten op, waaruit de volgende conclusies konden worden ge
trokken:
- De W -waarden blijven vrijwel constant, uitgezonderd bij plastic
buizen omhuld met esparto- of cocosvezel die een toeneming van de
intreeweerstand met de tijd vertonen.
317
- De beste combinatie bestaat uit gebakken drainbuizen met grind
(W = 2 dagen/m). e
Plastic ribbelbuizen met grind of gebakken buizen zonder grind
kunnen ook worden gebruikt (W = 5 dagen/m).
- Plastic ribbelbuizen zonder omhulling geven weinig bevredigende re-
sultaten (W = 1 3 dagen/m).
Plastic ribbelbuizen met een omhulling van esparto- of cocosvezel
bezitten zelfs een nog hogere intreeweerstand dan zonder omhulling.
- Gerststro is onbruikbaar als omhullingsmateriaal omdat het ge-
makkelijk rot en de buis verstopt.
Hoofdstuk 9
De ontzilting van het Alera proefveld begon met een periode van door-
spoeling, gevolgd door de verbouw van geirrigeerde, matig zoutresistente
gewassen.
De coefficient die het nuttig effect van de doorspoeling uitdrukt,
werd bepaald door een vergelijking van de werkelijke ontzilting met de,
volgens een theoretisch model, berekende ontzilting. Aldus was het mogelijk
de doorspoelingsbehoefte te voorspellen voor verschillende initiele zout-
gehalten.
Het zoutgehalte werd uitgedrukt in het chloridegehalte om de invloed
van slecht oplosbare zouten uit te schakelen. De correlatie tussen het chlo
ridegehalte en de elektrische geleidbaarheid was hoog. De uitkomsten van het
onderzoek leidden tot de volgende conclusies:
- De coefficient die het nuttig effect van de doorspoeling uitdrukt
was niet constant, maar nam met de diepte toe.
- De coefficient was hoger aan het begin van de ontzilting en nam
geleidelijk af naarmate de grond minder zout werd.
318
De berekende waarden van de coefficient weerspiegelden de verschil-
len in bodemstructuur.
Voor de bovenlaag (0 - 50 cm) werd een gemiddelde waarde van 0,5
bepaald en voor de volgende laag (50 - 100 cm) een waarde van 1.
Het zoutgehalte bij het begin hing af van bodemfysische eigenschappen,
zoals infiltratiesnelheid en doorlatendheid die op hun beurt weer
afhankelijk waren van de verdichting van de grond.
Bij een beginwaarde van 15 mmhos/cm voor de elektrische geleidbaar-
heid van de grond zijn ongeveer 1000 mm water voor doorspoeling no-
dig, hetgeen betekende een hoeveelheid irrigatiewater van 1100 tot
1400 mm. De doorspoelingsperiode kon 8 maanden duren, vanaf het be
gin van de herfst tot eind mei.
Diepwoelen en plaatselijke gipsbemesting verbeterden de structuur
van de bovenlaag.
De doorspoeling van zoute gronden vond in twee fasen plaats, namelijk
een eerste fase met alleen doorspoeling, gevolgd door een tweede fase
met de verbouw van een geirrigeerd, matig zoutresistent gewas (suiker-
bieten), waarbij de percolatieverliezen de diepere lagen doorspoelen.
Gedurende de doorspoelingsperiode bestond er geen gevaar voor alkali-
nisatie van de grond.
Om secondaire verzouting na ontginning te voorkomen, moeten de voor-
waarden voor een goede drainage gehandhaafd worden en dienen geirri-
geerde gewassen verbouwd te worden.
Hoofdstuk 10
De drainage criteria voor niet-stationaire omstandigheden werden afge-
leid uit het verband tussen grondwaterstand, gewasopbrengst en bewerkbaar-
heid van de grond, en uit het grondwaterstandsverloop in de winter en tijdens
de irrigatieperiode. Deze criteria werden vervolgens herleid tot criteria
319
voor stationaire omstandigheden die handiger zijn voor gebruik in drainage
projecten. Uit dit onderzoek werden de volgende conclusies getrokken:
- Wintergranen ondervinden weinig schade wanneer de grondwaterspiegel
zich niet langer dan drie opeenvolgende dagen op minder dan 50 cm
beneden maaiveld bevindt.
Bij grondwaterstanden tussen 75 en 100 cm beneden maaiveld vertonen
suikerbieten een goede groei en zij ondervinden geen schade indien
de waterstand tijdelijk, gedurende 3 tot 4 dagen, hoger komt dan
50 cm beneden maaiveld.
- Luzerne is gevoeliger voor hoge grondwaterstanden dan suikerbieten.
Voor goede opbrengsten dient de grondwaterstand niet langer dan 3
dagen op minder dan 25 cm beneden maaiveld, 4 tot 5 dagen op minder
dan 50 cm, en 5 tot 6 dagen op minder dan 75 cm beneden maaiveld
te staan.
- Grondwaterstanden minder dan 65 cm beneden maaiveld verhinderen het
gebruik van machines en belemmeren het bewerken van de grond,
zoals het gereedmaken van het zaaibed.
De volgende drainage criteria werden vastgesteld:
a) Wanneer in de winter de grondwaterspiegel tot aan maaiveld stijgt
moet deze in 8 dagen tot een diepte van 65 cm verlaagd worden.
b) Tijdens de irrigatieperiode moet de grondwaterstand, na een stij-
ging van 70 cm veroorzaakt door irrigatieverliezen, in de 12 dagen tussen
twee opeenvolgende irrigaties tot het oorspronkelijke niveau verlaagd worden
en na 7 dagen dieper dan 70 cm beneden maaiveld zijn.
Bij toepassing van deze criteria in de Boussinesq vergelijking voor
niet-stationaire stroming en door gebruik te maken van de doorlatendheid en
bergingscoefficient die op het proefveld waren bepaald, werd een afstand
320
van 25 m berekend voor drains op een diepte van 1,20 m. De hiermee overeen-
komende drainage criteria voor stationaire stroming waren een minimum diepte
van 0,5 m voor de onverzadigde zone, hetgeen leidt tot een potentiaal van
0,7 m en een drainafvoer van 3 mm/dag.
Wanneer rekening wordt gehouden met de intreeweerstand, konden de
vergelijking van Ernst voor stationaire stroming en die van Hellinga en
de Zeeuw voor niet-stationaire stroming worden gebruikt voor de berekening
van de drainafstand. Beide methoden leverden overeenkomstige resultaten op
waaruit de volgende conclusies konden worden getrokken:
- Wanneer drainage- en filtermaterialen met een hoge intreeweerstand
worden gebruikt zal de dichtheid van het netwerk van drains (in m/ha) twee-
maal zo groot zijn als bij materialen met een lage intree weerstand.
- Het gebruik van materialen met een hoge intreeweerstand houdt een
veel groter risico van mislukking in dan een wijde drainafstand met goed
materiaal.
321
Curriculum vitae
Julian Martinez Bettran was born in Soria (Spain) on September 19th
1947. In 1964, after completing secondary school at the Instituto "Antonio
Machado" in Soria, he studied at the Polytechnical University of Madrid,
graduating as Agricultural Engineer on December 9th 1969. From January to
June 1970 and from October 1972 to June 1973, he attended postgraduate
courses on Soil Science at the Madrid University.
In October 1970, he worked with the Ebro branch of the Instituto
Nacional de Colonizacion in soil surveys for irrigation projects. In June
1972 he joined the Soils Department of the Instituto Nacional de Reforma
y Desarrollo Agrario (IRYDA) in Madrid.
From August to December 1973 he attended the 12th International Course
on Land Drainage, held in Wageningen and organized by the International
Institute for Land Reclamation and Improvement (ILRI). After returning to
Spain, he began the research work for this thesis under the supervision
of Prof.Dr.W.H.van der Molen and Prof.Dr.P.Buringh of the University of
Agriculture, Wageningen. Field work was done at the Alera and Valarena
experimental fields in the Bardenas area and research work at the Department
of Land and Water Use and the Department of Tropical Soil Science of the
Wageningen University. The project was financed by IRYDA.
His present address is: Soils Department of IRYDA, Velazquez 147,
Madrid 2, Spain.
322